BOX ligands in biomimetic copper-mediated dioxygen activation: A hemocyanin model

FULL PAPER

DOI:10.1002/ejic.201402378

BOX Ligands in Biomimetic Copper-Mediated Dioxygen

Activation: A Hemocyanin Model

CLUSTER

ISSUE

Adam Walli,^[a]Sebastian Dechert,^[a]Matthias Bauer,^[b]

Serhiy Demeshko,^[a]and Franc Meyer*^[a]

Keywords: Bioinorganic chemistry / Copper / Enzyme catalysis / O-O activation / Peroxo complexes

The μ-η²:η²-peroxodicopper(II) core found in the oxy forms

of the active sites of type III dicopper proteins have been a

key target for bioinorganic model studies. Here, it is shown

that simple bis(oxazoline)s (BOXs), which are classified

among the so-called “privileged ligands”, provide a suitable

scaffold for supporting such biomimetic copper/dioxygen

chemistry. Three derivatives ^R,HBOX-Me₂(R = H, Me, tBu)

with different backbone substituents have been used. Their

bis(oxazoline)-copper(I) complexes bind dioxygen to yield

in solution and in the solid state. Resonance Raman spec-

troscopy revealed a characteristic isotope-sensitive stretch

ν

˜

= 731–742 cm^–1(Δ[¹⁸O₂] ≈ –39 cm^–1) and an intense fea-

O–O

ture around 280 cm^–1diagnostic for the fundamental sym-

metric Cu₂O₂core vibration. A slight butterfly-shape of the

Cu₂O₂core has been derived from EXAFS data and DFT cal-

culations. SQUID magnetic data evidenced strong antiferro-

magnetic coupled Cu^II₂(–2J Ն 1000 cm^–1). Thermal degrada-

tioninsolutionyieldsbis(hydroxo)-bridged[(^tBu,HBOX-Me₂)(L)-

biomimetic μ-η²:η²-peroxodicopper(II) species. O₂can be re- Cu(OH)]₂(PF₆)₂(L = H₂O, MeCN or THF); whereas in the

versibly released upon an increase in temperature. Their for- case of ^H,HBOX-Me₂, ligand oxygenation has been detected.

mation kinetics have been studied under cryo-stopped-flow

conditions for the tBu derivative, giving activation param-

Preliminary reactivity studies with the substrate 2,4-di-tert-

butylphenol indicate the formation of the C–C coupling prod-

uct 3,3Ј,5,5Ј-tetra-tert-butyl-2,2Ј-biphenol, whereas ortho-

hydroxylation was not observed. The copper(I) complex

eters

ΔH^‡

=

(2.27Ϯ0.18) kcalmol^–1

,

ΔS^‡

=

on

(–46.3Ϯ0.8) calK^–1mol^–1for the binding event and ΔH^‡

off

(11.7Ϯ1.9) kcalmol^–1, ΔS^‡= (–16.1Ϯ8.2) calK^–1mol^–1for [(^tBu,HBOX-Me₂)Cu(MeCN)]PF₆as well as two dicopper(II)

off

the release of O₂, as well as thermodynamic parameters ΔH°

complexes [(L)(^tBu,HBOX-Me₂)Cu(OH)]₂(PF₆)₂have been

characterised by single-crystal X-ray diffraction. Considering

=

(–10.0Ϯ1.7) kcalmol^–1

and ΔS° (–32.7Ϯ7.4)

=

calK^–1mol^–1for this equilibrium. The μ-η²:η²-peroxodicop- the vast number of known BOX derivatives, a rich and versa-

per(II) complexes have been isolated as surprisingly stable

solids and investigated in depth by a variety of methods, both

tile Cu/O₂chemistry based on this platform is anticipated.

Cu₂O₂motifs, have also been prepared.^[1,12–16]Most of

them are quite unstable even at low temperatures, whereas

others show reversible O₂-binding related to hemo-

cyanin.^[17–19]The particular Cu₂O₂structure obtained de-

pends on the precise nature of the capping ligands, inter

alia on their denticity and steric demand, and on the types

of counterion or solvent.^[1]Subtleties of the electronic struc-

tures of such Cu₂O₂cores have been studied in detail and

have been found to critically determine their spectroscopic

properties and reactivity.^{[1,2,10,20–24]}Hence, there is con-

tinuous interest in new and tuneable ligands and their re-

spective copper complexes that might support biomimetic

copper-dioxygen chemistry, and that might advance our un-

derstanding of the delicate subtleties of copper–dioxygen

interactions.

Introduction

Copper-dioxygen adducts have attracted much interest

during the last two decades because of their biological rele-

vance and their promising prospects for bioinspired oxi-

dation and oxygenation catalysis.^[1–4]Type III copper pro-

teins such as the O₂-transporter hemocyanin^[5,6]and the re-

lated enzymes tyrosinase^[7,8]and catechol oxidase^[9]contain,

in their oxy form, a side-on μ-η²:η²-peroxodicopper(II) core

that results from O₂-binding to the dicopper(I) deoxy state

(Figure 1).^[4,10,11]A number of synthetic analogue com-

plexes featuring the μ-η²:η²-peroxo unit (see also Table S1

in the Supporting Information), as well as other related

[a] Institut für Anorganische Chemie, Georg-August-Universität

Göttingen, Germany

Bis(oxazoline) derivatives (BOXs) are classified among

the so-called “privileged ligands” and have found wide-

spread applications,^[27–34]but surprisingly have not been ex-

plored in copper-mediated O₂activation until now. Having

recently developed some new BOX derivatives,^[35]we report

here a detailed investigation on the reversible O₂binding by

Tammannstr. 4, 37077 Göttingen

E-mail: franc.meyer@chemie.uni-goettingen.de

http://ww w .meyer.chemie.uni-goettingen.de/

[b] Department of Chemistry, University of Paderborn,

Warburger Str. 100, 33098 Paderborn, Germany

Supporting information for this article is available on the

WWW under http://dx.doi.org/10.1002/ejic.201402378.

Eur. J. Inorg. Chem. 2014, 4660–4676

4660

www.eurjic.org

FULL PAPER

diffraction analysis of a colourless single crystal obtained

from THF/Et₂O solution under inert conditions (Figure 2,

Table 1). The copper ion is coordinated in a trigonal planar

fashion by the ^tBu,HBOX-Me₂ligand with a bite angle N–

Cu–N of 93.6° and by an exogenous MeCN solvent mo-

lecule. The BOX ligand adopts a boat-shaped coordination

geometry. Owing to the steric bulk of the tBu group, the

angle between this residue and the {N₃Cu} coordination

plane is quite large (ca. 80°). NMR analysis of Cu^ItBu and

Cu^IH is in agreement with the solid-state structure being

maintained in solution. Cyclic voltammetry of a MeCN

solution of Cu^ItBu (see the Supporting Information and

Figures S1 and S2 for details) shows a quasi-reversible elec-

Figure 1. (Top) Schematic drawing of O₂binding to give the μ-

η²:η²-peroxodicopper(II) core. (Bottom) View of the active site of

the 73 kDa arthropod subunit II hemocyanin (Hc) of the Atlantic

horseshoe crab, Limulus polyphemus, determined by X-ray diffrac-

tion. Deoxy-Hc (left) at 2.2 Å resolution (PDB ID: 1LLA),^[25]

Cu···Cu distance: 4.61 Å; Oxy-Hc [right, μ-η²:η²-peroxodicop-

per(II)] at 2.4 Å resolution (PDB ID: 1OXY),^[26]distances [Å]:

Cu···Cu: 3.59, O–O: 1.41, Cu–N_eq: 2.1, Cu–N_ax: 2.4. Colour code:

white, protein backbone; blue, nitrogen; red, oxygen; yellow, cop-

per(I); teal, copper(II).

trochemical process with anodic peak potential E_pa

=

0.63 V vs. SCE, assigned to the Cu^IǞCu^II+ e^–oxidation.

Large separation of the anodic and cathodic peak poten-

tials (ΔE_p= 0.30 V) for this couple on the cyclic voltamme-

try time-scale is common for copper complexes because of

substantial structural rearrangements upon shuttling be-

tween Cu^Iand Cu^II, which favour very different coordina-

tion geometries.

copper(I) complexes of the three ligands ^R,HBOX-Me₂(R

= H, Me, tBu; Scheme 1 and Scheme 2) and on the resulting

dicopper–peroxo complexes.

Scheme 1. Preparation of copper(I)–bis(oxazoline) complexes Cu^IR

(R = tBu, Me, H).

Figure 2. X-ray solid-state molecular structure of Cu^ItBu. ORTEP

plot with significant atoms labelled. Displacement ellipsoids are

drawn at the 50% probability level; most hydrogen atoms have been

omitted for clarity; atoms located at a second site (B) due to crys-

tallographic disorder are drawn in lighter shade. See Figure S18

for van-der-Waals representations. Significant interatomic distances

and angles are listed in Table 1.

Scheme 2. Activation of dioxygen with copper(I)–bis(oxazoline)

complexes Cu^IR (R = tBu, Me, H) to form μ-η²:η²-peroxodicop-

per(II) complexes ^RP.

Table 1. Significant geometric information of the solid-state struc-

ture of Cu^ItBu in Figure 2. Interatomic distances [Å] and angles

[°].

Atoms 1,2

d 1,2 /Å

Atoms 1,2,3

Angle 1,2,3

Results and Discussion

Cu1–N1

Cu1–N2

Cu1–N3

N3–C16

1.9630(17)

2.0037(18)

1.8649(19)

1.131(3)

N1–Cu1–N2

N3–Cu1–N1

N3–Cu1–N2

93.56(7)

142.62(8)

123.64(8)

Copper(I) Complexes of BOXs

Cu1–N3–C16 175.0(2)

Reactions of the three ligands ^R,HBOX-Me₂(R = tBu,

Me, H) with [Cu(MeCN)₄]PF₆(Scheme 1) in tetra-

hydrofuran (THF) and other solvents give the air-sensitive

copper(I)

complexes

[(^R,HBOX-Me₂)Cu(MeCN)]PF₆

Activation of Dioxygen

Oxygenation of the colourless copper(I) complexes Cu^IR

(Cu^IR) as colourless solutions. For further reactions, the

complexes were generally not isolated, but their solutions

were used in situ. However, to establish their identity, at –78 °C in THF affords violet to purple μ-η²:η²-peroxodi-

Cu^ItBu and Cu^IH have been isolated and characterised, and copper(II) complexes ^RP (Scheme 2) with intense optical

the molecular structure of Cu^ItBu was ascertained by X-ray absorption bands at 333 nm (ε = 48.0Ϯ1.1 mm^–1cm^–1) and

Eur. J. Inorg. Chem. 2014, 4660–4676

4661

www.eurjic.org

FULL PAPER

Figure 3. UV/Vis spectra and formation of μ-η²:η²-peroxo complexes ^RP (left side, magenta lines), obtained upon oxygenation of Cu^IR

(black lines) in THF at –78 °C. R = (a) tBu, (b) H, and (c) Me. Final spectra after precipitation of peroxo complexes (blue lines, R = H,

Me). Concentrations of Cu^IR were 0.30 mm (R = tBu), 1.13 mm (R = H) and 0.79 mm (R = Me). Kinetic traces at 333 nm (right side,

circles); traces were obtained from an additional measurement (R = H) or from a cryo-stopped-flow experiment (R = tBu, with kinetic

first-order-rate fit, red line). See Table 2 for rate constants of formation and spectroscopic parameters.

500 nm (2.1Ϯ0.2 mm^–1cm^–1) in the case of ^tBuP and nearly π_σ*Ǟd_x₂_–y₂(in-plane) and π_v*Ǟd_x₂_–y₂(out-of-plane)

identical features in the case of ^HP and ^MeP (Figure 3, Cu^IIǟO_peroxoligand-to-metal charge transfer (LMCT)

Table 2). The oxygenation could be achieved either by the transitions, the former being more intense due to a better

injection of O₂gas into solutions of Cu^IR in THF (at overlap of orbitals.^[22,36]Although similar to absorptions of

–78 °C under Ar or N₂atmosphere) or by the slow injection previously reported systems, with λ_max(ε, mm^–1cm^–1) =

of the Cu^IR/THF solutions into O₂-saturated THF at ca. 340–380 nm (ca. 18–25) and ca. 510–550 nm (ca. 1),^[1]

–78 °C.

both CT bands of ^RP are at relatively high energy compared

The optical features are archetypical for a μ-η²:η²-per- to known systems and are remarkably more intense (indi-

oxo-dicopper(II) species and are closely analogous to the cated also by a Beer’s law plot for ^tBuP, see Figure S4). The

spectroscopic features of oxyhemocyanine and oxytyrosin- strong 333 nm transition for ^tBuP (ε = 48 mm^–1cm^–1) is only

ase. Prominent bands have previously been ascribed to comparable to model complexes with ethylene diamine de-

Eur. J. Inorg. Chem. 2014, 4660–4676

4662

www.eurjic.org

FULL PAPER

Table 2. Spectroscopic features and kinetic data of μ-η²:η²-peroxo- troscopy (λ_ex= 632.8 nm) of a solution of ^tBuP in THF

dicopper(II) complexes ^RP.

showed a Raman shift of 729 cm^–1(Figure 4), typical for

the weak ν(O–O) stretch of the μ-η²:η²-bound peroxide. An

intense feature at 278 cm^–1is also diagnostic for this side-on

peroxo complex and has been assigned as the fundamental

symmetric Cu₂O₂core vibration, ν(Cu–Cu).^[41]Additional

RR spectra are discussed below; no ¹⁸O isotopic substitu-

tion is provided for samples in solution due to problematic

solubility and weak intensity.

[c]

Solvent UV/Vis: λ_max[nm]

(ε [mm^–1cm^–1])^[a]

RR: ν [cm^–1]

k_obs

˜

(Δ[¹⁸O₂])^[b]

[min^–1]

Cu–Cu O–O

^tBuP THF

acetone 334 (47), 496 (1.1)

333 (48.0), 509 (2.1)

278

729

0.25

0.23

solid

THF

solid

263, 334, 520, ~ 620 (sh)

279 (1)

282 (0)

282

731 (39)

742 (39)

735

^HP

330 (Ͼ 30), 504 (Ͼ 0.8)^[d]

271, 332, 512, ~ 620 (sh)

333 (Ͼ 22), 496 (Ͼ 0.5)^[d]

261, 331, 493, 600^[e]

Ͼ 1

^MeP THF

solid

~ 1.5

[a] Solutions at –78 °C, solids (diffuse reflectance) at room temp.,

sh = shoulder. [b] Resonance Raman spectra of solutions at –60 °C,

solids at room temp. [c] Formation upon O₂injection at –78 °C.

[d] Low solubility. [e] Partly decomposed.

rived ligand systems (ca. 34–40 mm^–1cm^–1). Dicopper–per-

oxo complexes of those bidentate ligands usually feature

undefined weak axial solvent or counterion ligation; inti-

mate interaction with the couteranion was observed spec-

troscopically.^[37–39]This suggested the presence of weakly

bound axial solvent ligands also in ^RP, which is consistent

with EXAFS data (see below).

Figure 4. Resonance Raman (RR) spectrum of ^tBuP (λ_ex

=

632.8 nm) in THF solution at 213 K. The asterisk at 745 cm^–1indi-

–

cates PF₆.

The reactions of all three Cu^IR with O₂were found to

follow first-order kinetics under the applied conditions, and

the traces of the ~ 333 and ~ 500 nm features could be fitted

with single exponential functions to obtain rate constants

k_obs(Table 2). The rate constants were found to increase

with reduced steric demand of the backbone residue R. The

formation kinetics were investigated more closely with Cu-

^ItBu (see below). Whereas ^tBuP is both, soluble in THF and

sufficiently stable at low temperatures for an extended time,

the solubilities of ^HP and ^MeP were rather limited and they

were found to precipitate from their solutions during the

oxygenation reaction (Figure 3). Although the relatively

limited solubility of all three complexes is problematic to

analysis and experiments in solution, it is advantageous for

isolation of the peroxo complexes as solids (see below). The

UV/Vis bands of ^tBuP are similar in acetone and in THF

solutions (Table 2, Figure S6a), whereas no peroxo complex

formation was observable in CH₂Cl₂.

Cryo-Stopped-Flow Kinetic Measurements

The kinetics of O₂activation by Cu^ItBu were studied by

using stopped-flow-UV/Vis techniques under cryo condi-

tions (see Experimental section and the Supporting Infor-

mation for details). Analysis of all recorded UV/Vis spectra,

from stopped-flow and other experiments, showed no de-

tectable accumulation of intermediates. This suggests the

following mechanism, as likewise found in related com-

plexes:^[2]

Here, the initially formed CuO₂complex is instantly

trapped by a second Cu^ItBu molecule and the first step is

the rate-determining step. With an excess of O₂present, this

reaction follows pseudo-first-order conditions:^[42,43]

^RP are thermolabile at temperatures above approximately

–60 °C. The oxygenation reaction of ^tBuP was found to be

reversible in THF (Figure S5): warming to –15 °C under

k_obs= k_on·[O₂] + k_off

Variation of the O₂concentration led to a second-order

inert gas, followed by recooling and admitting O₂, allowed rate constant k_onfor O₂binding and a first-order constant

sequential cycles of O₂binding and release. The half-life k_offfor the release, from the plot of k_obsvs. O₂concentra-

t_1/2of ^tBuP upon warming to –15 °C was around 3 min. Ap- tion (Figure 5, Table 3). The plot shows a linear relationship

plying vacuum to a solution of ^tBuP in acetone induces O₂with a relatively large y-intercept above –60 °C. Such a ki-

release already at –78 °C; however, slow decomposition of netic behaviour is clearly indicative for a reversible O₂-bind-

^tBuP was evident in acetone (Figure S6).

ing step, hence an equilibrium Cu^ItBu + O₂i [CuO₂] is

In μ-η²:η²-peroxodicopper(II) complexes, the O–O bond indicated. Below –60 °C, the intercept is close to zero, re-

is significantly weakened compared with H₂O₂(ν

=

flecting an equilibrium that is shifted to the right

˜

O–O

880 cm^–1) due to backbonding; known systems have charac- (k_on·[O₂] ӷ k_off) with the oxygenated species being predomi-

teristic resonance Raman (RR) frequencies and isotope nant in the reaction mixture. However, at temperatures

shifts of ν_O–O= 730–760 cm^–1(Δ[¹⁸O₂] ca. 40 cm^–1). In con- above –60 °C, from the y-intercept, k_offcan be obtained in

˜

trast, 1,2-trans-peroxo, 1,2-cis-peroxo,^[40]and bis(μ-oxo) di- addition to k_on. Reversibility is also confirmed by the ex-

copper systems have ν

≈ 830 cm^–1(46), ν_O–O≈ 800 cm^–1periments described above, demonstrating that O₂is re-

˜

O–O

(45) and ν

≈ 600 cm^–1(28), respectively.^[1]RR spec- moved by applying vacuum or gentle warming, and that the

˜

Cu–O

Eur. J. Inorg. Chem. 2014, 4660–4676

4663

www.eurjic.org

FULL PAPER

Table 3. Kinetic (k_on, k_off) and thermodynamic (K_eq) parameters for the formation of 1:1 adduct [CuO₂] upon low-temperature oxygena-

tion of Cu^ItBu in THF (obtained from cryo-stopped-flow experiments and from the plots in Figure 6).

Activation parameters

Binding

T [K]

k_on[m^–1s^–1]

ΔH^‡[kcalmol^–1]

ΔS^‡[calK^–1mol^–1]

on

233

223

213

2.69Ϯ0.06

2.19Ϯ0.42

1.54Ϯ0.18

2.27Ϯ0.18

–46.3Ϯ0.8

Dissociation

T [K]

k_off[s^–1]

ΔH^‡[kcalmol^–1]

ΔS^‡[calK^–1mol^–1]

off

233

223

213

(1.58Ϯ0.02)ϫ10^–2

(5.6Ϯ1.5)ϫ10^–3

(7.9Ϯ6.5)ϫ10^–4

11.7Ϯ1.9

–16.1Ϯ8.2

Thermodynamic parameters

T [K]

[a]

K_eq[m^–1]

ΔH° [kcalmol^–1]

–10.0Ϯ1.7

ΔS° [calK^–1mol^–1]

–32.7Ϯ7.4

233

223

213

(1.70Ϯ0.04)ϫ10²

(3.9Ϯ1.3)ϫ10²

(2.0Ϯ1.6)ϫ10³

[a] K_eq= [CuO₂]/([Cu^ItBu]·[O₂]) = k_on/k_off

.

^tBuP complex is then reconstituted by repeated exposure to oxygen carrier enzymes have significantly larger binding

O₂. In addition, both rate constants of the equilibrium al- constants at room temperature (e.g., hemocyanin,

low determination of the equilibrium constants K_eq5.7ϫ10⁵m^–1).^[45]

(Table 3).

Figure 5. Plots of k_obsvs. concentration of dioxygen at –40, –50

and –60 °C for the activation of dioxygen with Cu^ItBu in THF;

data from cryo-stopped-flow measurements.

Activation and equilibrium parameters (Table 3) were de-

termined from Eyring and van’t Hoff plots, respectively Figure 6. Eyring plots, ln(k T^–1) vs. T^–1, of (a) O₂binding (k_on) and

(b) dissociation (k_off) steps with weighted linear regressions. (c) The

(Figure 6). The system features a favourable reaction en-

van’t Hoff plot, ln(K_eq) vs. T^–1, for the equilibrium Cu^ItBu + O₂

thalpy of ΔH° = (–10.0Ϯ1.7) kcalmol^–1and an unfavour-

i [CuO₂] {K_eq= [CuO₂]/([Cu^ItBu]·[O₂]) = k_on/k_off}. See Table 3 for

derived kinetic and thermodynamic parameters.

able reaction entropy of ΔS° = (–32.7Ϯ7.4) calK^–1mol^–1.

The thermodynamics determined for 1:1 adduct formation

are similar to most copper(I)/dioxygen systems, which have

Reported activation enthalpies for oxygenation of cop-

ΔH°

≈

–7 to –10 kcalmol^–1and ΔS°

≈

–26 to per(I) complexes are in a quite narrow range of ΔH^‡≈ 5–

on

–33 calK^–1mol^–1.^[2]In comparison, Cu^ItBu is at the lower 10 kcalmol^–1. Unfavourable activation entropies are gen-

edge of those ranges for both parameters. The small en- erally observed, and are mostly in the range ΔS^‡_on≈ –14 to

thalpy suggests strong dioxygen binding, which parallels he- 2 calK^–1mol^–1, consistent with an associative mechanism;

mocyanin and other biological dioxygen carriers. In con- that is, loss of degrees of freedom upon binding of O₂. In

trast, the unfavourable entropy of formation hampers the comparison to reported systems, for Cu^ItBu an un-

observation of the copper–oxygen adducts at room tem- usually small activation enthalpy of ΔH^‡

=

on

perature in the case of low-molecular-weight systems.^[44](2.27Ϯ0.18) kcalmol^–1was found. This suggests that vac-

Therefore, the equilibrium constant for Cu^ItBu is rather ant coordination sites are present in Cu^ItBu, and that direct

small, even at –60 °C (2ϫ10³m^–1), whereas noncooperative attack of O₂can take place without any interference from

Eur. J. Inorg. Chem. 2014, 4660–4676

4664

www.eurjic.org

FULL PAPER

solvent (direct oxygenation);^[2,46]such a very low activation vent (a partial decomposition is additionally indicated by

energy is typical of diffusion-controlled processes (1– the methods described below).

4 kcal).^[47]No significant alterations of bonds seem to be

Although bidentate ligands suffice in stabilising Cu₂O₂

necessary to proceed to the transition state. The compara- species, it should be noted that an additional labile ligand

tively small enthalpy might be explained by the fact that (solvent or counteranion) is generally associated with each

Cu^ItBu is only twofold-coordinated by the BOX ligand, metal ion; in case of more common tridentate capping li-

whereas most reported systems, for which kinetic param- gands, one of the donors is usually weakly bound.^[1,39]This

eters were determined for the 1:1 Cu/O₂adduct formation, parallels the coordination of copper in the role model en-

contain tridentate capping ligands.

An associative mechanism is further reflected in the duces precipitation of the present complexes. However, the

strongly negative activation entropy of ΔS^‡

differences in coordinating ability towards transition metals

zymes (Figure 1). The coordination of THF presumably in-

=

on

(–46.3Ϯ0.8) calK^–1mol^–1. Because of this substantial and between MeCN (coordinating ability index^[51]a^TM= –0.2)

negative ΔS^‡, the rate of O₂binding to Cu^ItBu in THF and THF (–0.3) are not significant. When in solution, ^RP

at –78 °C (k_obs= 0.2 min^–1) is slow compared with other are stable only at very low temperatures (also reflected by

systems.^[2]This very unfavourable activation entropy can the strongly negative ΔS^‡), and – considering the thermola-

additionally be blamed for the high thermolability of this bility of Cu/O₂complexes in general – ^tBuP and ^HP are re-

system. A reason for this strongly negative entropy might markably insensitive in the solid state. Handling in air at

be the bidentate ligand and the vacant coordination side, room temperature for short times was straightforward;

with respect to coordinative unsaturation of the copper(I) however, storage for prolonged periods was only possible at

complex: no solvent dissociation is needed to take place low temperatures.

upon initial attack of O₂. The release of coordinated solvent

Diffuse reflectance UV/Vis spectroscopy of violet ^RP

might in fact entropically compensate for the association of powder samples shows absorption maxima in accordance

O₂to some extent in the other systems reported previously. with bands observed for the solutions (Figure 7, Table 2).

However, steric bulk around Cu, arising from the dimethyl The relative intensities of the absorption bands are ob-

groups on the oxazoline rings, might also play a role in the scured due to light scattering by the powders. The spectra

present case (cf. van-der-Waals plots of Cu^ItBu in Fig- exhibit two intense bands at ca. 330 nm and ca. 515 nm,

ure S18). The steric demand of the ligand was also sug- similar to those of the solution spectra. This suggests that

gested to be the reason for particularly unfavourable acti- the μ-η²:η²-peroxodicopper(II) structures are present in

vation entropies for O₂binding to other copper com- both, solution and powder, and it can be presumed that the

plexes,^[44,48]as well as to Fe^II{tris(pyridylmethyl)amine}^[49]spectra do not result from a mixture of chromophores. In

and Co^II(cyclidene)^[47]systems.

the case of ^MeP, for which partial decomposition was as-

The direct attack of O₂onto a free coordination site pre- sumed, the 330 nm feature is also consistent with the solu-

sumably leaves the MeCN bound to Cu in the CuO₂inter- tion spectrum and the other solid samples. However, a

mediate adduct, because threefold coordination would be 493 nm peak is at slightly higher energy and lower intensity,

uncommon for Cu^II. Inspection of the back reaction (k_offcompared with the ca. 515 nm peaks. Furthermore, an ad-

process) also reveals a strongly negative entropy, here ΔS^‡

ditional feature of similar intensity is present at 600 nm; a

off

= (–16.1Ϯ8.2) calK^–1mol^–1, indicating that the back reac- similar feature is only visible as a shoulder in the case of

tion also follows an associative mechanism. Here, associa-

tion of solvent molecules (MeCN or THF) to the CuO₂

complex is likely to be rate-limiting, rather than O₂-dissoci-

ation. A similar kinetic behaviour was observed for, e.g.,

the formation of a copper(II)–superoxo (1:1) complex.^[50]

Finally,

the

activation

enthalpy

ΔH^‡

=

off

(11.7Ϯ1.9) kcalmol^–1reflects a significant reaction barrier

for this solvent-association step.

Isolation and Analysis of Solid Peroxodicopper Compounds

The limited solubility of the ^RP complexes allowed their

isolation in good yields in the form of violet to pink pow-

ders (Figure S3). Elemental analysis indicates excellent to

satisfactory purity for ^tBuP and ^HP and confirms the formu-

lations as [{(^R,HBOX-Me₂)(THF)Cu}₂(O₂)](PF₆)₂, which

includes one coordinated THF molecule per copper ion.

For ^MeP, partial decomposition was clear from a slightly

greyish colour tint; the analysis showed a comparatively low

Figure 7. Solid-state UV/Vis/NIR spectra (diffuse reflectance) of

^tBuP and ^HP in KBr powder at room temp. The ^MeP sample reflects

some degree of degradation. Spectra are relatively scaled at 330 nm;

an artefact is present at the detector/grating change point of

800 nm. For comparison, the spectrum of ^tBuP in THF solution

(–78 °C) is depicted (Ͻ 800 nm). Spectral properties are compared

CHN content, in agreement with loss of coordinated sol- in Table 2.

Eur. J. Inorg. Chem. 2014, 4660–4676

4665

www.eurjic.org

FULL PAPER

^tBuP and ^HP. For other μ-η²:η²-peroxodicopper(II) com- 1sǞ3d pre-edge transitions at 8979.5 eV were found. Cu^III

plexes, where it was possible to isolate a solid sample and in related compounds is characterised by a feature of ca.

to record reflectance spectra, the spectra were also in good 2 eV higher energy,^[52,53]which is not observable in the sam-

agreement with solution data.^[12,18,36]

ples here. The 1sǞ4p + ligand-to-metal (LMT) shakedown

RR spectroscopy of the solid samples of ^RP shows the transitions at ca. 8987 eV are only present as weak shoul-

presence of μ-η²:η²-coordinated peroxide (Figure 8, ders with no significant peak intensity; this excludes an oxi-

Table 2). The respective ν(O–O) stretches, increasing with dation state of Cu^IIIthat is characterised by a sharp peak

the steric demand of the backbone residue, are detected at on the rising edge.^[52]Furthermore, Cu^Iis usually indicated

731 (^tBuP), 735 (^MeP, Figure S10) and 742 cm^–1(^HP); ν(Cu– by a diagnostic sharp and intense 1sǞ4p transition at

Cu) stretches are found at 278 (^tBuP) and 282 cm^–1(^HP, 8984 eV,^[4,54]with a shape that is dependent on the coordi-

^MeP). The features are essentially unchanged from the solu- nation environment; this feature is also absent in the spec-

tion data for ^tBuP (see above). Isotopic substitution, i.e., tra. Therefore, from the XANES results, a Cu^IIstate in all

oxygenation of the respective copper(I) complex with ¹⁸O₂, ^RP is concluded, in agreement with side-on μ-η²:η²-peroxo-

was carried out for ^tBuP and ^HP and effectively shifts the dicopper(II) complexes. Nonetheless, subtle differences be-

peroxo stretches to 692 and 703 cm^–1, respectively, in both tween the samples can be observed in the white-line shape.

cases with a shift of Δ[¹⁸O] = –39 cm^–1, whereas the ν(Cu– All samples exhibit a double white-line structure with sim-

Cu) features are essentially unchanged, as expected.^[41]In ilar intensity of both white-line components; however, the

–

^HP, the 742 cm^–1feature is obscured by the PF₆stretch intensity ratio of the two underlying signals differ. The

at 745 cm^–1, which, however, is well-separated in the ¹⁸O₂- white-line width of sample ^MeP is reduced compared with

labelled sample. The difference spectrum (Figure 8) finally those of ^HP and ^tBuP. The white-line signature of ^RP was

also unveils the ν(¹⁶O–¹⁶O) stretch. In addition, RR and IR similarly found for other binuclear μ-η²:η²-peroxodicopper

spectroscopy confirm the absence of MeCN and the pres- complexes.^[4,17,53]

–

ence of PF₆counterions in the solid samples (cf. Fig-

ures S8–S10), in accordance with elemental analysis.

Figure 9. Cu K-edge XANES spectra of ^tBuP, ^MeP and ^HP. The spec-

tra are shifted on the ordinate for clarity.

Metric parameters obtained from Cu K-edge extended

X-ray absorption fine-structure (EXAFS) data (Figure 10,

Table 4) are typical for μ-η²:η²-peroxodicopper(II) com-

plexes. For comparison, all μ-η²:η²-peroxodicopper com-

plexes that have been structurally characterised so far are

listed in Table S1.

In ^tBuP and ^HP, the first shell at 1.94 Å agrees well with

Figure 8. RR spectra of solid powder samples of (a) ^tBuP and

a fourfold coordination, which is an average value of the

(b) ^HP, with ¹⁶O¹⁶O (blue) and ¹⁸O¹⁸O (magenta) isotopic composi- two oxygen and two nitrogen atoms from peroxo and BOX

tion (632.8 nm laser excitation) and ¹⁶O₂minus ¹⁸O₂difference

ligands [Cu–(N/O)_eq]. A second light-atom contribution at

spectra (red). No ¹⁶O/¹⁸O isotope scrambling was detected.

2.34 (^tBuP) or 2.32 Å (^HP), with a coordination number

slightly smaller than one, is attributed to weakly coordinat-

ing axial THF (Cu–O_ax), in accordance with elemental

analysis data. These distances are in agreement with each

Structural Investigations

Cu ion having a square-pyramidal geometry. The Cu···Cu

Synchrotron-based X-ray absorption spectroscopy distance of (3.51Ϯ0.04) Å (^tBuP) and (3.52Ϯ0.04) Å (^HP)

(XAS) was performed on the ^RP powders. The X-ray ab- is characteristic for a μ-η²:η²-peroxo dicopper(II) complex,

sorption near-edge structure (XANES) spectrum estab- which was also confirmed by the Cu···Cu coordination

lished the Cu^IIoxidation states (Figure 9). Low intensity number of one, reflecting the dimeric nature of the complex.

Eur. J. Inorg. Chem. 2014, 4660–4676

4666

www.eurjic.org

FULL PAPER

scribed.^[17,54]The Cu···Cu distance and the corresponding

Cu–O–Cu multiscattering path (3.72 Å) are in agreement

with the Cu₂O₂core having a slightly butterfly-shaped ge-

ometry, which is confirmed by DFT calculations for ^tBuP

(see below). In contrast to these μ-η²:η²-peroxodicopper(II)

complexes, end-on 1,2-trans-peroxodicopper(II) complexes

would show a Cu···Cu distance of ca. 4.6 Å, whereas bis(μ-

oxo)dicopper(III) complexes are characterised by a short

Cu···Cu separation of ca. 2.80 Å.^[1]Furthermore, the ther-

mal degradation products of ^tBuP, bis(hydroxo)dicopper(II)

complexes, have Cu···Cu distances of 3.0 Å as determined

by X-ray crystallography (see below). None of those

Cu···Cu distances could be detected in ^HP and ^tBuP.

In contrast to ^tBuP and ^HP, no second light-atom contri-

bution could be adjusted for ^MeP in a statistically significant

manner. Instead, the first contribution exhibits a slightly

higher coordination number than in ^tBuP and ^HP, suggesting

that a solvent contribution cannot be resolved anymore, but

is averaged with the ligand contribution. In ^MeP, two

Cu···Cu shells were necessary to achieve a satisfactory qual-

ity of fit. The shorter one found at 2.93 Å can be interpre-

ted by the presence of a bis(μ-hydroxo)dicopper(II) species,

whereas the longer one (3.49 Å) is still in agreement with

the peroxo species found in the case of ^tBuP and ^HP. The

coordination numbers of 0.6 and 0.2 suggest that the per-

oxo species represents the minor fraction. It should also be

noted that the Cu···Cu coordination numbers do not add

to unity, therefore a small fraction of monomeric species

could be also present. The reduced fraction of peroxo spe-

cies is also reflected in the reduced Cu–O–Cu multiple scat-

tering contribution. However, the remaining presence of

Cu–NC multiple scattering indicates that the ligand scaffold

is still intact in the present species.

Figure 10. EXAFS spectra k³χ(k) (top) and the corresponding Fou-

rier transformed functions (bottom) of ^tBuP, ^MeP and ^HP. The ex-

perimental spectra are shown as black solid lines, the fitted spectra

according to the parameters given in Table 4 are shown as red

dashed lines. The spectra are shifted on the ordinate for clarity.

Table 4. Structural parameters obtained by fitting the experimental

EXAFS spectra in Figure 10 with theoretical models.

Sample Abs–Bs^[a]

N(Abs)^[b]R(Abs–Bs)^[c]σ^[d]

R^[e][%]

E_f^[f][eV]

[Å]

Theoretical Investigations

^tBuP

Cu–(N/O)_eq

Cu–O_ax

Cu···Cu

Cu–O–Cu_MS

Cu–NC_MS

Cu–(N/O)_eq

Cu–O_ax

4.6Ϯ0.5 1.94Ϯ0.02

0.7Ϯ0.1 2.34Ϯ0.02

1.0Ϯ0.3 3.51Ϯ0.04

3.5Ϯ0.7 3.72Ϯ0.04

9.0Ϯ1.8 4.28Ϯ0.04

4.8Ϯ0.5 1.94Ϯ0.02

0.5Ϯ0.1 2.32Ϯ0.02

0.9Ϯ0.3 3.52Ϯ0.04

3.9Ϯ0.7 3.73Ϯ0.04

7.8Ϯ1.6 4.29Ϯ0.04

5.4Ϯ0.5 1.94Ϯ0.02

0.6Ϯ0.1 2.93Ϯ0.03

0.2Ϯ0.1 3.49Ϯ0.04

1.6Ϯ0.7 3.79Ϯ0.04

9.9Ϯ2.0 4.29Ϯ0.04

0.071Ϯ0.007 22.48

0.100Ϯ0.010 8.2

0.095Ϯ0.020

0.102Ϯ0.020

0.071Ϯ0.007 19.71

0.100Ϯ0.010 7.4

0.087Ϯ0.020

0.102Ϯ0.020

0.100Ϯ0.020

0.074Ϯ0.007 20.33

0.100Ϯ0.010 7.9

0.032Ϯ0.010

0.084Ϯ0.017

0.107Ϯ0.022

Density functional theory (DFT) was applied to model

the structure of ^tBuP, with MeCN as coordinating solvent. A

DFT spin-unrestricted formalism has been applied to first

calculate a high-spin triplet state (S = 1) geometry, which

was then used to calculate the spin-polarized broken-sym-

metry (BS) solution.^[55,56]The spin on one copper atom was

flipped in expectation of an antiferromagnetically coupled

(M_S= 0) state. In the successfully obtained BS solution,

Mulliken spin populations at the copper atoms are +0.4930

and –0.4932. Figure 11 shows the obtained geometry and a

plot of the spin-density.

^HP

Cu···Cu

Cu–O–Cu_MS

Cu–NC_MS

Cu–(N/O)

Cu···Cu

Cu–Cu

Cu–O–Cu_MS

Cu–NC_MS

^MeP

UHF natural orbitals are depicted in Figure S14. The

computations reproduce well the experimental bond lengths

from EXAFS data within a mean error of ca. 2% (Table S2)

as well as the suggested slight butterfly-shape of the Cu₂O₂

core (Cu–O–O–Cu torsion angle is 152°); the nature of the

[a] Abs = X-ray absorbing atom, Bs = backscattering atom.

[b] Number of neighbour backscattering atoms. [c] Distance be-

tween Abs and Bs. [d] Debye–Waller-like factor. [e] Quality of fit.

[f] Shift between experimental and theoretical EXAFS function.

Surprisingly, Cu–C shells around 3 Å could not be adjusted HOMO of the butterfly structure is reminiscent of that pre-

with statistical significance, which indicates a rather high viously reported for such structure.^[57]The computed

degree of disorder in the solid samples. Higher shells at dis- Cu···Cu separation of 3.56 Å is only slightly overestimated

tances larger than 3.5 Å are caused by Cu–O–Cu and Cu– compared with the experimental distance of 3.51 Å derived

N–C multiscattering (MS) paths, that are similarly de- from EXAFS data.

Eur. J. Inorg. Chem. 2014, 4660–4676

4667

www.eurjic.org

FULL PAPER

coupling are both extremely high, –2J Ն 1000 cm^–1for ^tBuP

and –2J Ն 1200 cm^–1for ^HP, based on H = –2J·S₁·S₂(Fig-

ure 13). Above room temp., up to 330 K, a slight increase

of χ_molT is discernible in the case of solid ^HP (grey circles,

not considered in the fits). A subsequent measurement after

cooling to 295 K showed increased χ_molT compared with

the initial values for the pristine sample around this tem-

perature (grey square), suggesting that irreversible decom-

position of the complex occurred above room temp.

Figure 11. Broken symmetry (M_S= 0) DFT geometry-optimised

coordinates of ^tBuP (left) and spin-density plot (right, isovalue =

0.004). Colour code: Cu, teal; N, blue; O, red; C, grey; H, white.

B3LYP/def2-TZVPP (Cu,N,O)/def2-SVP(C,H) level of theory.

Using time-dependent (TD) DFT, the electronic spec-

trum of ^tBuP was successfully reproduced (Figure 12, see the

Supporting Information for details). The calculated spec-

trum closely corresponds with the experimental spectrum.

The intensity pattern for the UV/Vis absorptions agrees

with the experimental spectrum and with the typical

pattern^[1]for a μ-η²:η²-peroxodicopper(II) complex [higher

energy π_σ*(O₂^2–)ǞCu^IImore intense than lower energy

π_v*(O₂^2–)ǞCu^IItransition; intensities ca. 25:1].

Figure 13. Plots of χ_molT vs. T (circles) for (a) ^tBuP and (b) ^HP ob-

tained from SQUID measurements at 0.5 T. The red solid lines each

correspond to the best fit for two antiferromagnetically coupled S

= 1/2 ions with g = 2.0. Additional fit parameters: temperature-

independent paramagnetism [TIP of (a) 990ϫ10^–6and

(b) 185ϫ10^–6cm³mol^–1K] and paramagnetic impurity [grey

dashed lines, PI of (a) 4 and (b) 11.6 mol-% for magnetic uncou-

pled Cu^IIspecies as likely impurities]. The fits give lower limits

of the exchange coupling constant –J of (a) Ն 500 cm^–1and

(b) Ն 600 cm^–1. The blue lines represent simulations assuming

weaker coupling (smaller –J values). In (b), a change in the sample

Figure 12. UV/Vis absorption spectra of ^tBuP. Grey line: experimen-

tal low-temperature UV/Vis spectrum (λ_max= 333, 500 nm; THF,

–78 °C). Blue sticks: computed TD-DFT transitions. Red line: con-

volution of calculated transitions (full width at half maximum:

3000 cm^–1), which reproduces well the experimental solution spec-

trum (λ_max= 340, 520 nm); the intensities are scaled relatively.

occurred Ͼ 290 K (grey circles), indicated by

a subsequent

measurement at 295 K (grey square); due to this, the grey data

points were not considered in the fits.

An equally large singlet–triplet splitting of –2J Ն

600 cm^–1was found for oxyhemocyanine.^[58]Coupling con-

stants could also be determined for synthetic model systems

Magnetic Coupling

Examples of experimentally determined magnetic ex- in a few cases. Similar values were found for two μ-η²:η²-

change coupling constants for copper–dioxygen systems are peroxo bridged model systems, –2J Ն 600^[59]and –2J Ն

scarce,^{[36,40,58–60]}due to the thermal lability of most com- 800 cm^–1,^[36]as well as for a model compound with trans-

plexes. Usually, EPR silence is taken as an indication of a 1,2-peroxo-coordination (–2J Ն 600 cm^–1).^[59,60]Only re-

diamagnetic ground state, and “normal” NMR spectro- cently was a weakly coupled peroxo-dicopper(II) complex

scopic behaviour (i.e., no paramagnetically shifted reso- reported (–2J = 144 cm^–1), characterised by a cis-1,2-peroxo

nances) is taken as evidence for the presence of strong anti- binding mode and an acute Cu–O–O–Cu torsion of 65°.^[40]

ferromagnetic coupling. Magnetic susceptibility measure- Our DFT calculations for ^tBuP are in agreement with the

ments using a SQUID magnetometer confirm the S = 0 experimentally determined coupling (–2J ≈ 3000 cm^–1; see

ground state of solid ^tBuP and ^HP, resulting from very strong the Supporting Information for details), and are further-

antiferromagnetic coupling between the cupric ions. Simu- more in the range of calculations for other μ-η²:η²-peroxo-

lations of the data show that the lower limits of exchange dicopper(II) systems.^[61]

Eur. J. Inorg. Chem. 2014, 4660–4676

4668

www.eurjic.org

FULL PAPER

Thermal Degradation

After warming to room temp., solutions of intense violet

^tBuP bleach or turn bluish; the violet powder of ^tBuP also

turns light-blue after some time at room temp., which is

indicative of the formation of the corresponding bis-

(hydroxo)-bridged Cu₂(OH)₂complex as the decomposition

product. The decomposition was also clearly evident in the

Raman spectrum of this powder, which lacks the intense

archetypical ν(O–O) and ν(Cu–Cu) features of ^tBuP (Fig-

ure S11). The [(^tBu,HBOX-Me₂)₂Cu₂(OH)₂(PF₆)]⁺ion was

identified in the high-resolution ESI mass spectrum (Fig-

ure S12). A solution of ¹⁸O-labelled ^tBuP at room temp. af-

forded the corresponding ¹⁸O-labelled bis(hydroxo) com-

plex. However, approximately 50% ¹⁶O/¹⁸O isotope scram-

bling was evident, and it is expected that an exchange with

water from solvents or air is relatively fast. Similarly, the

IR spectrum of a powder sample, prepared by leaving ¹⁸O-

labelled ^tBuP at room temp. and in air for some days,

showed no difference in OH stretch compared to a sample

with normal isotopic composition.

Single crystals of [(^tBu,HBOX-Me₂)(L)Cu(OH)]₂(PF₆)₂

could be isolated from two different solutions and were

analysed by X-ray diffraction (L is either THF/MeCN or

H₂O; Figure 14). Metric parameters of the central

Cu^II₂(OH)₂units (Table 5), such as the Cu···Cu separation,

are similar to those reported for other Cu^II₂(OH)₂di-

mers.^[62]The complex resides on a crystallographically im-

posed inversion centre; the bulky tBu ligand residues are

located on opposite sides of the Cu₂(OH)₂core. All cop-

per(II) ions are found in square-pyramidal environment (τ

= 0.02 and 0.10, respectively).^[63]In analogy to ^tBuP, solvent

ligands are weakly bound (2.2–2.3 Å) in the axial positions.

We found no indication for any oxidative degradation of

the ^tBu,HBOX-Me₂ligand itself. This is important, consider-

Figure 14. X-ray solid-state molecular structures of [(H₂O)-

(

^tBu,HBOX-Me₂)Cu(OH)]₂(PF₆)₂(top) and [(THF)_0.85(MeCN)_0.15

-

^tBu,HBOX-Me₂)Cu(OH)]₂(PF₆)₂(bottom). ORTEP plots with sig-

nificant atoms labelled. Displacement ellipsoids are drawn at the

50% probability level; anions and most hydrogen atoms have been

omitted for clarity; MeCN located at a second site due to crystallo-

graphic disorder are drawn in lighter shade; symmetry transforma-

tions used to generate equivalent atoms (Ј) in each complex: 1 – x,

1 – y, –z. See Figure S19 for further representations. Significant

interatomic distances and angles are listed in Table 5.

Table 5. Selected atom distances and bond angles for [(L)(^tBu,HBOX-Me₂)Cu(OH)]₂(PF₆)₂with L = THF/MeCN or H₂O, shown in

Figure 14.

Atoms 1,2

L

d 1,2 [Å]

THF/MeCN

Atoms 1,2,3

L

Angle 1,2,3 [°]

THF/MeCN

H₂O

Cu1···Cu1Ј

O3···O3Ј

Cu1–N1

Cu1–N2

Cu1–O3

Cu1–O3Ј

Cu1–O4

Cu1–N3

3.0435(6)

2.406(2)

1.979(2)

1.986(2)

1.9376(16)

1.9420(17)

2.288(4)

2.35(4)

2.9968(7)

2.450(3)

1.981(3)

N1–Cu1–N2

O3–Cu1–O3Ј

N1–Cu1–O3Ј

N2–Cu1–O3

N1–Cu1–O3

N2–Cu1–O3Ј

N1–Cu1–O4

N2–Cu1–O4

O3–Cu1–O4

O3Ј–Cu1–O4

N1–Cu1–N3

N2–Cu1–N3

O3–Cu1–N3

O3Ј–Cu1–N3

O4–Cu1–N3^[a]

Cu1–O3–Cu1Ј

91.65(10)

76.65(7)

94.54(9)

95.16(8)

166.15(9)

167.64(8)

94.40(12)

92.35(10)

97.35(10)

97.83(9)

99.7(8)

101.5(7)

90.8(8)

88.1(7)

10.7(6)

103.35(7)

91.82(10)

78.54(8)

93.75(9)

93.85(9)

163.65(9)

169.71(9)

100.08(10)

95.87(9)

94.58(9)

91.67(9)

1.986(2)

1.9348(17)

1.9361(17)

2.246(2)

101.46(8)

φ(Cu1–O3–O3Ј–Cu1Ј)

τ value^[b]

180

0.02

180

0.10

[a] Note that this angle is between two disordered molecules. [b] τ = (β – α)/60°;^[63]trigonal bipyramidal, τ = 1; square pyramidal, τ = 0.

Eur. J. Inorg. Chem. 2014, 4660–4676

4669

www.eurjic.org

FULL PAPER

ing that intramolecular ligand oxygenation or oxidation has try can be expected because numerous BOX derivatives are

often been observed in Cu₂O₂complexes with other cap- readily available.

ping ligands.^[2,64]

No analogous Cu₂(OH)₂complex has been isolated from

^HP so far. Experimental evidence suggests, however, that the

Experimental Section

ligand is oxygenated at the backbone upon degradation of

Materials and Instrumentation: Unless otherwise stated, all solvents

used were of commercially available analytical, spectroscopic or

HPLC grade. Diethyl ether was dried by heating to reflux with

sodium/benzophenone, tetrahydrofuran (THF) with potassium/

^HP. When dissolved in CH₂Cl₂or acetone at room temp.,

solutions of ^HP turn red to brown. Analysis by ESI mass

spectrometry indicated the presence of oxygenated ligand.

The observed base peak (m/z 225.1, [^OBOX-Me₂+ H]⁺)

benzophenone and both were distilled under an argon atmosphere.

matches the pattern expected for the oxygenated ligand with

bridging C=O as the backbone. It was so far possible to

isolate only a small amount of oxygenated ligand ^OBOX-

Methylene chloride (CH₂Cl₂) was degassed and then dried by a

series of drying columns in an MBraun solvent purification system

(MB-SPS 800). Acetone was degassed and dried by stirring over

boron trioxide (B₂O₃) for 24 h with a desiccant loading of 5% w/v

and then distilled under argon.^[66]All solvents were kept in storage

flasks with an attached J. Young PTFE valve under an atmosphere

of dry argon. Argon 5.0 (99.999%) was passed through a column

of SICAPENT (phosphorus pentoxide on inert carrier). Prepara-

tion and handling of air-sensitive compounds were performed un-

der an argon atmosphere using standard Schlenk techniques or in

an MBraun LABmaster SP inert atmosphere glovebox workstation,

filled with nitrogen with glovebox atmosphere levels of oxygen and

moisture Ͻ 0.1 ppm. Deoxygenation of solvents and solutions was

effected by either sparging with argon for 15–20 min or repeated

freeze–pump–thaw cycles. All chemicals were purchased from com-

mon distributers and were used without further purification. Di-

oxygen 4.5 (99.995%) was from Linde; isotopically enriched di-

oxygen-18 (¹⁸O₂) was from Campro Scientific and was of 99.7%

purity with the isotopic composition ¹⁸O/¹⁷O/¹⁶O of 97.1:0.9:2.0

atom-%. Tetrakis(acetonitrile)copper(I) hexafluorophosphate, [Cu-

1

Me₂. Analysis by H and ¹³C NMR spectroscopy proved

the absence of any H atoms at the C atom linking the two

oxazoline rings, and also a corresponding cross-peak signal

was absent in an ¹H,¹³C-HSQC experiment. Such ligand

oxygenation reactivity is likely, when considering the closely

related diketiminate (NacNac) ligand class. It has pre-

viously been shown that a NacNac ligand is oxygenated in

an analogous fashion by heating to 50 °C in air for 12 h

with Cu(OAc)₂.^[65]Further investigations are ongoing to

unravel mechanistic details of the ligand oxygenation upon

thermal degradation of ^HP.

Reactivity Towards External Substrates

^tBuP was used for preliminary reactivity studies because

the stability and solubility of ^MeP and ^HP are limited. The

results showed that ^tBuP reacts with 2,4-di-tert-butylphenol

to give the C–C coupling product 3,3Ј,5,5Ј-tetra-tert-butyl-

2,2Ј-biphenol, whereas ortho-hydroxylation was not ob-

served. Although there may be electronic reasons associated

with the butterfly-type core,^[57]it should be noted that ^tBuP

is quite compact due to the bulky dimethyl and tert-butyl

substituents, as well as the bound solvent molecules [cf. van-

der-Waals plots of the bis(hydroxo)dicopper(II) complex in

Figure S19]. Thus, the central Cu₂O₂core seems to be diffi-

cult to access for large substrates. Further reactivity studies

are ongoing.

(MeCN)₄](PF₆), was synthesised by following

a

reported

method.^[67]The ligands ^R,HBOX-Me₂(R = H, Me, tBu) were pre-

pared by following the procedures we reported recently.^[35]

NMR spectra were recorded at room temperature (unless otherwise

noted) with a Bruker Avance DRX 500 MHz, Avance 300 MHz or

400 MHz spectrometer. ¹H and ¹³C chemical shifts (δ) are reported

in ppm relative to SiMe₄and were determined relative to the resid-

ual solvent peaks. ¹⁵N, ³¹P and ¹⁹F chemical shifts are reported in

ppm relative to nitromethane, phosphoric acid and trichlorofluoro-

methane respectively, and were referenced externally using the sol-

vents deuterium lock frequency. Multiplicities are indicated [br

(broadened), s (singlet), d (doublet), sept (septet)]. ¹⁵N NMR reso-

nances were recorded by using ¹H,¹⁵N-HMBC experiments. As-

1

signment of all H and ¹³C NMR resonances was possible, using

¹H,¹³C-HSQC and ¹H,¹³C-HMBC experiments when necessary.

Electrospray mass spectra (ESI-MS) were recorded with an Applied

Biosystems API 2000 spectrometer or a Bruker HCTultra (high-

capacity ion trap) spectrometer with an additional inlet connected

to an MBraun glovebox, and capable of further fragmentation of

ions (MS²). ESI^+/–indicates that the detector was used either in

Conclusions

Cu^ItBu was shown to reversibly bind dioxygen and to

serve as a new functional hemocyanin model. The resulting

μ-η²:η²-peroxodicopper(II) complex ^tBuP has been charac- positive or negative mode. High-resolution mass spectra (HRMS)

were recorded with a Bruker APEX IV spectrometer (Fourier

Transform Ion Cyclotron Resonance, FTICR-ESI-MS) or a Bruker

micrOTOF (time of flight, TOF-ESI-MS) spectrometer. High-reso-

lution ions and isotopic patterns were simulated with Bruker Com-

pass DataAnalysis 4.0. Vibrational (IR) spectra were recorded with

a Digilab Excalibur Series FTS 3000. Samples were ground with

potassium bromide and pressed to a pellet. Peaks are reported with

the following intensities: s (strong), m (medium), w (weak) and br

(broad); ν_asand ν_sindicate asymmetric and symmetric stretching

modes respectively. Resonance Raman (RR) spectra were measured

terised by a variety of methods, both in solution and in

the solid state, and thermodynamic and kinetic parameters

relating to its formation have been determined. Studies on

the two analogous μ-η²:η²-peroxodicopper(II) complexes

^HP and ^MeP support the results, but these compounds exhi-

bit more pronounced thermal lability and limited solubility,

which is inherent to this class. This work demonstrates that

simple bis(oxazoline) derivatives, which represent a privi-

leged ligand class, are suitable scaffolds for supporting bio-

mimetic Cu/O₂species. A rich and versatile Cu/O₂chemis- with a Horiba LabRAM HR 800 Raman spectrometer equipped

Eur. J. Inorg. Chem. 2014, 4660–4676

4670

www.eurjic.org

FULL PAPER

with an open-electrode CCD detector using a backscattering micro-

scope optic. A He/Ne mixed-gas laser was used for excitation; spec-

tra were recorded with an excitation wavelength λ_ex= 632.8 nm.

Thermally unstable samples were measured in a CryoVac Konti-

Cryostat-Mikro at 213 K, cooled by the flow of liquid N₂-cooled

N₂gas. Solutions of ^RP were generated directly in the measuring

cell by oxygenation with O₂gas. Solid samples were measured in

air at ambient temperature. Isotopic substitution was achieved by

oxygenating with ¹⁸O₂gas. The recorded spectra were processed

with Horiba LabSpec 5 software. The compositions of C, H and N

were determined with an Elementar 4.1 vario EL 3 elemental ana-

lyser and were carried out by the Analytical Laboratory of the In-

stitute of Inorganic Chemistry at the University of Göttingen. UV/

Vis/NIR electronic spectra were recorded at room temperature with

a Varian Cary 50 Bio spectrophotometer. Low-temperature mea-

surements were performed directly in the reaction solution with an

all-quartz immersion probe in a custom-made reaction tube (trans-

mission measurement with 1 mm optical path, Hellma Analytics).

The spectra were recorded by the use of a fibre-optical connection

and analysed by Cary Win UV and OriginLab OriginPro 8.5.1 soft-

ware. The temperature was maintained with a dry ice in acetone

cooling bath (–78 °C) or a dry ice in ethylene glycol cooling bath

(ca. –20 °C). Oxygenation of the copper(I) complex was achieved

by introduction of molecular dioxygen into cooled reaction solu-

tions by bubbling through a stainless steel syringe needle, or by

slow injection of complex solution into an oxygen saturated and

cooled solution. Subsequent removal of excess dioxygen was car-

ried out by bubbling with argon and/or repeated vacuum–argon

cycles. Extinction coefficients (ε) are uncorrected for solvent con-

traction (THF ca. 10%, from ambient temperature to 195 K). Pow-

der samples were measured with a Varian Cary 5000 instrument by

using a Harrick Praying Mantis diffuse reflection accessory. Sam-

ples were ground with potassium bromide and measured in a

microsampling cup at room temperature; sh stands for an absorp-

tion shoulder.

2CHH), 4.21 (d, J = 8.9 Hz, 2 H, 2CHH), 3.29 (s, 1 H, CH), 2.28

(s, 3 H, MeCN), 1.39 (s, 6 H, 2CH₃), 1.38 (s, 6 H, 2CH₃), 1.10 (s,

9 H, tBu) ppm. ¹³C{¹H} NMR (101 MHz, CD₂Cl₂): δ = 168.0

(C=N), 80.3 (CH₂), 68.1 (CMe₂), 48.5 (CH), 37.0 (CMe₃), 28.8

(2Me), 28.7 (CMe₃), 28.6 (2Me), 3.2 (MeCN) ppm, MeCN was not

observed. ¹⁵N NMR (30 MHz, CD₂Cl₂): δ = –161.4 (C=N) ppm,

MeCN was not observed. UV/Vis (THF or acetone): no absorption

(300–800 nm).

[(^H,HBOX-Me₂)Cu(MeCN)]PF₆(Cu^IH): ^H,HBOX-Me₂(63 mg,

0.30 mmol) was dissolved in anhydrous THF (40 mL) under an at-

mosphere of argon in a small Schlenk flask. The solution was

sparged for ca. 15 min with argon to ensure the absence of oxygen.

[Cu(MeCN)₄]PF₆(112 mg, 0.30 mmol, 1.00 equiv.) was added and

the obtained colourless solution was stirred for ca. 15 min. The

solvent was removed in vacuo and the complex was obtained quan-

titatively after drying in vacuo as an air-sensitive white powder. ¹H

NMR (300 MHz, CHCl₃): δ = 4.20 (s, 4 H, oxazoline-CH₂), 3.47

(s, 2 H, bridging-CH₂), 2.34 (s, 3 H, MeCN), 1.41 (s, 12 H, BOX-

CH₃) ppm. ¹³C{¹H} NMR (75 MHz, CDCl₃): δ = 79.7 (oxazoline-

CH₂), 67.8 (CMe₂), 28.3 (Me), 26.6 (bridging-CH₂), 2.8

(MeCN) ppm, C=N and MeCN were not observed. ³¹P NMR

1

(121 MHz, CDCl₃): δ = –145 (sept, J_P,F= 714 Hz, PF₆) ppm. ¹⁹F

1

NMR (282 MHz, CDCl₃): δ = –73 (d, J_F,P= 713 Hz, PF₆) ppm.

UV/Vis (THF): no absorbance (300–800 nm).

[(^Me,HBOX-Me₂)Cu(MeCN)]PF₆(Cu^IMe): The complex was pre-

pared in situ and not isolated. UV/Vis (THF): no absorbance (300–

800 nm).

Preparation and Isolation of [{(THF)(^tBu,HBOX-Me₂)Cu}₂-

(O₂^2–)](PF₆)₂^tBuP): ^tBu,HBOX-Me₂(67 mg, 0.25 mmol) was dis-

(

solved in anhydrous THF (50 mL) in a Schlenk flask and degassed

by bubbling with argon to make a 5.0 mm solution. [Cu(MeCN)₄]-

(PF₆) (93 mg, 0.26 mmol, 1.0 equiv.) was added and a second

Schlenk flask, filled with dry O₂was linked to the first flask

through a short glass adapter filled with glass-wool or through a

glass frit. The solution was then cooled to –78 °C and the appara-

tus was transferred to a freezer and stored at –82 °C. Upon slow

diffusion of O₂into the copper(I) complex solution, a violet pre-

cipitate (cf. Figure S3) appeared. The colourless supernatant was

carefully removed by using a stainless steel syringe needle under

an argon atmosphere and the residue was thoroughly washed with

anhydrous diethyl ether; this was repeated three times. The residue

was carefully dried for a short period in vacuo to yield ^tBuP (86 mg,

61%) as a very fine violet powder. The powder could be warmed

to room temperature, without any problems with respect to decom-

position, over at least some hours, and could be stored in air at

–80 °C for at least several months. C₃₈H₆₈Cu₂F₁₂N₄O₈P₂: calcd. C

40.53, H 6.09, N 4.98; found C 40.52, H 6.13, N 4.91. UV/Vis

(THF, –78 °C): λ_max(ε, m^–1cm^–1) = 333 (48 000), 500 (2100) nm.

UV/Vis (powder; diffuse reflection, r. t.): λ_max= 263, 334, 520,

[(^tBu,HBOX-Me₂)Cu(MeCN)]PF₆(Cu^ItBu). Method 1: ^tBu,HBOX-

Me₂(50 mg, 188 μmol, 1.05 equiv.) was dissolved in anhydrous

CH₂Cl₂(40 mL) under an atmosphere of argon. The solution was

sparged for ca. 15 min with argon to ensure the absence of oxygen.

[Cu(MeCN)₄]PF₆(66 mg, 170 μmol, 1.00 equiv.) was added and the

mixture was stirred for 30 min until all salt was dissolved. A colour-

less solution was obtained. The solvent was removed in vacuo and

the resulting residue was heated to 80 °C under vacuum to remove

excess ligand to obtain Cu^IBOX quantitatively as an air-sensitive

white powder.

Method 2: ^tBu,HBOX-Me₂(24 mg, 90 μmol) was dissolved in an-

hydrous THF (5 mL) under an atmosphere of argon in a small

Schlenk flask. The solution was sparged for ca. 15 min with argon

to ensure the absence of oxygen. [Cu(MeCN)₄]PF₆(34 mg,

90 μmol, 1.0 equiv.) was added and the obtained colourless solution

was stirred for some minutes. The flask was then linked to a second

Schlenk flask, containing anhydrous, degassed diethyl ether

(40 mL), by a bent glass adapter. The slow gaseous diffusion of

Et₂O into the THF solution afforded small crystalline white rods

of the copper(I) complex after a few days. The single crystals ob-

tained were of high quality and were suitable for X-ray structure

determination. The supernatant was removed and the residue was

thoroughly washed with anhydrous degassed Et₂O (5ϫ 4 mL) un-

der argon by using a syringe with a stainless steel needle. After

drying in vacuo, Cu^ItBu (12 mg, 26%) was obtained as a white

crystalline solid. MS (ESI⁺, MeCN): m/z (%) = 595.3 (17) [2 BOX

ca. 620sh nm. RR (powder, r.t., λ_ex= 632.8 nm, selected peaks): ν

˜

= 1657 (ν{C=N}), 745 (ν_s{PF₆}), 731 (ν{O–O}), 279 (ν{Cu–

Cu}) cm^–1. IR (KBr, r.t.): ν = 2975 (s, ν{C–H}), 2940 (m, ν{C–

˜

H}), 2880 (m, ν{C–H}), 1650 (vs, ν{C=N}), 1475 (m), 1566 (m),

1438 (s), 1407 (w), 1388 (m), 1374 (m), 1324 (m), 1283 (w), 1259

(w), 1198 (s), 1179 (s), 1161 (m), 1063 (s), 1051 (m), 1030 (w), 1017

(w), 966 (s), 840 (vs, br, ν_as{PF₆}), 788 (m), 744 (m, ν_s{PF₆}), 665

(w), 558 (vs, PF₆), 521 (w) cm^–1. IR in Nujol mull provided iden-

tical peaks.

¹⁸O₂isotopic substitution was achieved by leaving ¹⁸O₂(20 mL,

1 atm), under an atmosphere of argon and at –82 °C, slowly diffuse

into a solution of Cu^IBOX in THF (5.0 mm, 10 mL, 50 μmol). Af-

+ Cu]⁺, 370.0 (100) [(BOX)Cu(MeCN)]⁺, 329.0 (32) [(BOX)-

1

Cu]⁺. H NMR (400 MHz, CD₂Cl₂): δ = 4.25 (d, J = 8.9 Hz, 2 H, ter 1 d, the colourless supernatant was removed by using a syringe

Eur. J. Inorg. Chem. 2014, 4660–4676

4671

www.eurjic.org

FULL PAPER

and the violet precipitate was washed with anhydrous Et₂O (5 mL). [(H₂O)(^tBu,HBOX-Me₂)Cu(μ-OH)]₂(PF₆)₂: Powder of ^tBuP was dis-

After careful drying in vacuo, ¹⁸O₂-labelled ^tBuP (26 mg, 92%) was

obtained. RR (powder, r.t., λ_ex= 632.8 nm, selected peaks): ν =

solved in chloroform at room temperature. From this solution, a

small number of turquoise crystals crystallised by slow concentra-

˜

1657 (ν{C=N}), 745 (ν_s{PF₆}), 692 (ν{O–O}), 278 (ν{Cu– tion in air. A crystal that was suitable for X-ray diffraction analysis

Cu}) cm^–1.

was obtained from this solution. C₃₀H₅₈Cu₂F₁₂N₄O₈P₂·3H₂O

calcd. C 33.55, H 6.01, N 5.22: found C 33.64, H 5.40, N 5.20. IR

Preparation and Isolation of [{(THF)(^H,HBOX-Me₂)Cu}₂-

(O₂^2–)](PF₆)₂(^HP): ^H,HBOX-Me₂(53 mg, 0.25 mmol) was dissolved

in anhydrous THF (50 mL) and the solution was sparged with ar-

gon for 30 min. [Cu(MeCN)₄]PF₆(93 mg, 0.25 mmol, 1.00 equiv.)

was added and the mixture was stirred until full dissolution. The

solution was cooled to –78 °C and O₂gas was generously injected.

The solution turned slowly deep violet, and was stirred for 14 h.

The colourless supernatant was carefully removed by using a stain-

less steel needle and the fine violet precipitate was washed with

Et₂O (2ϫ 10 mL). The residue was suspended in a small amount

of Et₂O and the suspension was transferred to a small flask. The

supernatant was removed and the residue was carefully dried in

vacuo for a short period. The complex (96 mg 95 μmol, 76%) was

stored at –80 °C without any sign of decomposition as very fine,

deep-pink powder. C₃₀H₅₂Cu₂F₁₂N₄O₈P₂: calcd. C 35.54, H 5.17,

N 5.53; found C 34.97, H 5.09, N 5.38. UV/Vis (THF, –78 °C):

λ_max(ε, m^–1cm^–1) = 330 (Ͼ 30 000), 504 (Ͼ 800) nm. UV/Vis (pow-

der, diffuse reflection, r.t.): λ_max= 271, 332, 512, ca. 620 (sh) nm.

(Nujol mull, selected peaks): ν = 3649 (w, ν{O–H}), 3583 (w, ν{O–

˜

H}), 1654 (m, ν{C=N}), 844 (s, vs, br, ν_as{PF₆}), 741 (m, ν_s{PF₆}),

558 (vs, PF₆) cm^–1.

Thermal Degradation of ^HP: ^HP (30 mg, 26 μmol) was dissolved in

CH₂Cl₂(10 mL) at room temp. The solution turned red and, after

some minutes, green. The solution was stirred overnight, then 5%

NH₄OH solution (20 mL) was added and the mixture was vigor-

ously stirred for several minutes. The aqueous phase turned slowly

blue and was discarded. This was repeated two times. The organic

phase was separated and washed with H₂O (2ϫ 20 mL) and dried

with Na₂SO₄. The solvent was removed to obtain a small amount

of residue. The residue was extracted with a small amount of

CDCl₃and the solution was filtered. MS (ESI⁺, MeCN): m/z (%)

1

= 225.1 (100) [[^OBOX-Me₂+ H]⁺]. H NMR (300 MHz, CDCl₃):

δ = 3.97 (s, 4 H, CH₂), 1.29 (s, 12 H, Me) ppm. ¹³C{¹H} NMR

(75.5 MHz, CDCl₃): δ = 79.4 (CH₂), 68.5 (CMe₂), 28.0 (Me) ppm,

C=N and C=O could not be observed. ¹⁵N (30.4 MHz): δ = –220.5

(C=N) ppm.

RR (powder, r.t., λ_ex= 632.8 nm, selected peaks): ν = 745 (ν PF ),

˜

s

6

742 (ν{O–O}), 282 (ν{Cu–Cu}) cm^–1.

Reactivity of ^tBuP Towards 2,4-Di-tert-butylphenol (DTBP): Anhy-

drous THF (8.0 mL) was cooled to –78 °C and the solution was

saturated with dioxygen by injection over a period of 10 min. A

5.0 mm solution of Cu^ItBu in THF (2.0 mL) was slowly injected

into the cool solution; slow formation of ^tBuP (0.5 mm) was evident

by the slow change from colourless to intense violet. After stirring

for 15 min, triethylamine (10 equiv., 7.0 μL, 50 μmol) was added,

followed by the addition of DTBP in fivefold excess (25 μmol). The

solution was stirred at –78 °C for 16 h, along with slow warming

of the cooling bath to room temperature. No change in colour of

the solution was evident after some hours at –78 °C. The solvent

was removed in vacuo and the residue was redissolved in dichloro-

methane (5 mL). HCl solution (6 m, 5 mL) was added, the phases

were separated and the aqueous phase was extracted with dichloro-

methane (3ϫ 10 mL). The organic phases were combined and the

volatiles were removed in vacuo. The residue was dried in vacuo

and analysed by NMR spectroscopy. The residue contained the un-

changed phenol (62%), and 38% was converted into the C–C cou-

pling product 3,3Ј,5,5Ј-tetra-tert-butyl-2,2Ј-biphenol (TBBP); no

3,5-di-tert-butyl-o-quinone (DTBQ) was detected. ¹H NMR

(300 MHz, CDCl₃): δ = 7.38 (d, J = 2.3 Hz, 2 H, Ar_TBBP), 7.26 (d,

J = 2.5 Hz, 1 H, Ar_DTBP), 7.10 (d, J = 2.5 Hz, 2 H, Ar_TBBP), 7.04

(dd, J = 8.1, 2.1 Hz, 1 H, Ar_DTBP), 6.75 (d, J = 8.2 Hz, 1 H,

Ar_DTBP), 1.44 (s, 18 H, tBu_TBBP), 1.40 (s, 9 H, tBu_DTBP), 1.31 (s,

18 H, tBu_TBBP), 1.28 (s, 9 H, tBu_DTBP) ppm.

¹⁸O₂isotopic substitution was achieved in an analogous reaction;

but divergently, ¹⁸O₂gas (10 mL, 1 atm) in argon was allowed to

diffuse slowly into the Cu^IH/THF solution from a second Schlenk

flask through a bent glass joint at –80 °C, yield 38 mg (37 μmol,

30%); very fine, deep-pink powder. RR (powder, r.t., λ_ex

=

632.8 nm, selected peaks): ν = 745 (ν PF ), 703 (ν{¹⁸O–¹⁸O}), 282

˜

s

6

(ν{Cu–Cu}) cm^–1.

Preparation of [{(^Me,HBOX-Me₂)Cu}₂(O₂^2–)](PF₆)₂(^MeP): The com-

plex was prepared from ^Me,HBOX-Me₂(56 mg, 0.25 mmol) and

[Cu(MeCN)₄]PF₆(93 mg, 0.25 mmol), by following the method

used for ^tBuP. However, the complex was found to decompose partly

during isolation, yield 68 mg; very fine, pink to grey powder.

C₂₄H₄₀Cu₂F₁₂N₄O₆P₂clacd. C 32.11, H 4.49, N 6.24; found C

32.01, H 4.67, N 5.54. UV/Vis (THF, –78 °C): λ_max(ε, m^–1cm^–1) =

333 (Ͼ 22 000), 496 (Ͼ 500) nm. UV/Vis (powder, diffuse reflec-

tion, r.t.): λ_max= 261, 331, 493, ca. 600 nm. RR (powder, r.t., λ_ex

=

632.8 nm, selected peaks): ν = 745 (ν PF ), 735 (ν{O–O}), 282

˜

s

6

(ν{Cu–Cu}) cm^–1.

Thermal Degradation of ^tBuP and Preparation of [L(^tBu,HBOX-Me₂)-

Cu(μ-OH)]₂(PF₆)₂: Storage of powdered ^tBuP at room temp. led

slowly to a colour change from intense violet to a light-blue, which

was indicative of degradation of the peroxo complex and formation

of the corresponding bis(μ-hydroxo)dicopper(II) complex. Corre-

spondingly, solutions turned slightly blue at room temperature.

[(THF)_0.85(MeCN)_0.15

(

^tBu,HBOX-Me₂)Cu(μ-OH)]₂(PF₆)₂:

A

solu-

Beer’s Law Plots for ^tBuP: Solutions of varying concentrations of

^tBuP (0.1–0.3 mm) in THF were prepared by injection of Cu^ItBu

solutions into chilled, saturated solutions of O₂in THF in a dry

ice/acetone bath; stock solutions of Cu^ItBu (2 and 5 mm) were

used. The solutions slowly became purple, and the development of

the full spectrum (250–700 nm) was monitored with a 0.1 cm path

tion of ^tBuP in THF at –78 °C was warmed to r.t., which was ac-

companied by decolouration. Under an atmosphere of argon, a

small number of turquoise crystals grew from this solution after a

few days, which were suitable for X-ray crystal structure determi-

nation. MS (ESI^–, MeCN): m/z (%) = 145 (100) [PF₆]. HRMS

–

(ESI⁺, MeCN): m/z (%) = 837.2287 (3) {[BOX₂Cu₂(OH)₂PF₆]⁺; length immersion probe until no change was observed. The slopes

C₃₀H₅₄Cu₂F₆N₄O₆P requires 837.2272}, 346.1313 (100) {[(BOX)- of the absorbances at 333 and 500 nm versus concentration of ^tBuP

Cu(OH)]⁺; C₁₅H₂₇CuN₂O₃requires 346.1312}, 329.1285 (83)

give extinction coefficients of approximately 48 000 and

{[(BOX)Cu]⁺C₁₅H₂₆CuN₂O₂requires 329.1284}. IR (Nujol mull, 2 100 m^–1cm^–1, respectively. These values are not corrected for the

selected peaks): ν = 3648 (m, ν{O–H}), 3579 (m, ν{O–H}), 2264

ca. 10% solvent contraction from room temp. to –78 °C. Concen-

trations of 0.3 mm or higher caused precipitation of ^tBuP over the

short or long term.

˜

(w, ν_MeCN{CϵN}), 1650 (vs, ν{C=N}), 839 (vs, br, ν_as{PF₆}), 742

(m, ν_s{PF₆}), 557 (vs, PF₆) cm^–1.

Eur. J. Inorg. Chem. 2014, 4660–4676

4672

www.eurjic.org

FULL PAPER

Reversibility of O₂Activation by ^tBuP: A solution of ^tBuP was pre-

pared as described above by the O₂injection method in an acetone/

sation of exchange coupling and Zeeman splitting was performed

with the julX program.^[71]Temperature-independent paramagne-

dry ice bath at –78 °C; its formation was monitored by UV/Vis tism (TIP) and a Curie-behaved paramagnetic impurity (PI) with

spectroscopic analysis. Excess O₂was then removed by several vac-

uum/argon cycles. The flask was warmed to –15 °C by using an

ethylene glycol/dry ice bath. Reoxygenation was achieved by ad-

ditional exchange of cooling baths and repeated injection of O₂.

spin S = 1/2 were included according to χ_calcd.= χ·(1 – PI)

+ χ_mono·PI + TIP. The coupling constant J is related to the Heisen-

berg–Dirac–van-Vleck Hamiltonian in the form of H = –2J·S₁·S₂.

X-ray Absorption Spectroscopy (XAS, XANES, EXAFS): Measure-

ments were performed at the XAS beamline of the ANKA synchro-

tron (Karlsruhe, Germany) under ambient conditions at 293 K. A

Si(111) double crystal monochromator was used for measurements

at the Cu K-edge (8.979 keV). The second monochromator crystal

was tilted for optimal harmonic rejection. The spectra were re-

corded in transmission mode with ionisation chambers filled with

nitrogen. The individual pressures were adjusted to optimise the

signal to noise ratio. Energy calibration was performed with a cop-

per metal foil. The solid samples were embedded in an oxygen-

free boron nitride matrix and pressed into pellets. Data evaluation

started with background absorption removal from the experimental

absorption spectrum by subtracting a Victoreen-type polyno-

mial.^[72,73]To determine the smooth part of the spectrum, corrected

for pre-edge absorption, a piecewise polynomial was used. It was

adjusted in such a way that the low-R components of the resulting

Fourier transform were minimal. After division of the background-

subtracted spectrum by its smooth part, the photon energy was

converted into photoelectron wavenumbers k. The resulting χ(k)-

function was weighted with k³and Fourier transformed using a

Hanning window function. Data analysis was performed in k-space

on unfiltered data. Adjustment of the common theoretical EXAFS

expression

Electrochemical Measurements: Cyclic voltammetry (CV) and

squarewave voltammetry (SWV) were conducted with Cu^ItBu in a

0.1 m solution of tetrabutylammonium hexafluorophosphate

(TBAHP, [NBu₄]PF₆; electrochemical grade) as the supporting

electrolyte in MeCN solution. Measurements were performed at

room temp. with a potentiostat/galvanostat Perkin–Elmer Model

263A, controlled by Electrochemistry Power Suite software. A

glassy carbon electrode (milli-GC) was used as the working elec-

trode, platinum as counter electrode and silver as reference elec-

trode. All solutions were deoxygenated by bubbling with argon and

measurements were performed under anaerobic conditions. The de-

camethylferrocene/-ferrocenium couple, [Fe(Cp*)₂]^+1/0{E_1/2

=

–0.59 V vs. ferrocene, [Fe(Cp)₂]^+1/0, in MeCN}^[68,69]was added after

the measurements as the internal standard, to avoid the potential

overlap of the Fe(Cp)₂redox signal with other edox processes. The

potential was finally converted into E vs. SCE (saturated calomel

electrode) by adding 0.40 V.

Cryo-Stopped-Flow Kinetics: Measurements were performed with a

single mixing BioLogic Science Instruments μSFM-20 Stopped-

Flow Module with two syringes, controlled by a BioLogic MPS-20

Microprocessor Unit. Temperature was regulated with a Huber CC

75 cryostat with the observation head immersed into the cryosolv-

ent (DW-Therm, DWS Synthesetechnik). The observation head

was connected by an umbilical link to the μSFM-20. Spectra were

recorded with an optical fibre-mounted photodiode array spectro-

photometer J & M Analytik AG TIDAS. The temperature in the

observation head was observed with an internal probe. The as-

sembly was controlled with Bio-Kine32 software. The stopped-flow

apparatus was flushed with argon-saturated THF to ensure the ab-

sence of air. Saturated solutions of O₂in THF were prepared by

bubbling dry O₂gas through anhydrous THF for ca. 15 min. The

solubility of O₂in THF at 25 °C and 1 atm is reported to be

10.0 mm (molar fraction χ = 8.16ϫ10^–4),^[70]this value was used in

the calculation of the O₂concentrations. Since the system was

closed and the solutions had no contact to the gas phase, no change

in concentration was evident. No correction for the temperature

contraction of the solvent was applied. The reaction kinetic profiles

were deconvoluted with the SPECFIT/32 global analysis software.

Pseudo-first-order rate constants k_obswere obtained from nonlin-

ear regression of spectral traces with single exponential function

using OriginLab OriginPro 8.5.1 software. The k_onand k_offvalues

were obtained from a linear regression of a plot of k_obsvs. O₂con-

centration. Activation and equilibrium parameters were obtained

from linear regressions of the respective Eyring and van’t Hoff plots

using instrumental weighted fitting with OriginPro.

(N_j: one type of neighbour atoms j in a shell, r_j: distance of atoms

2

j from the X-ray absorbing atom, S₀: amplitude reduction factor,

F_j: backscattering amplitude, σ²: Debye–Waller like factor, δ_j: over-

all phaseshift) according to the curved wave formalism of the EX-

CURV98 program with XALPHA phase and amplitude func-

tions^[74]yielded the structural parameters given in Table 4. The

mean free path of the scattered electrons was calculated from the

imaginary part of the potential (VPI set to –4.00 eV). Since an am-

plitude reduction factor was used to account for inelastic pro-

cesses,^[75]it was determined experimentally from the reference com-

pounds Cu₂O, CuO and Cu(OH)₂to be 0.8Ϯ0.1 by setting the

coordination numbers to the crystallographic values.^[76–78]An inner

potential correction E_fwas introduced when fitting experimental

data with theoretical models, which accounts for an overall phase

shift between the experimental and calculated spectra. In the fitting

procedure, it was taken into account that the number of fitted pa-

rameters (N_pars) did not exceed the degrees of freedom (N_ind) that

are calculated according to N_ind= (2ΔkΔR/π).^[79]The quality of fit

is given in terms of the R-factor according to:^[80]

Temperature-Dependent Magnetic Susceptibility: Measurements

were carried out with a Quantum Design MPMS-XL-5 supercon-

ducting quantum interference device (SQUID) magnetometer

equipped with a 5 Tesla magnet in the temperature range from 295

to 2 K. The powder sample was contained in a Teflon bucket and

fixed in a nonmagnetic sample holder. Each raw data file for the

measured magnetic moment was corrected for the diamagnetic con-

tribution of the sample holder and the Teflon bucket. The molar

susceptibility data was corrected for the diamagnetic contribution.

Simulation of the experimental data with full-matrix diagonali-

X-ray Crystallography: X-ray data were collected with a STOE

IPDS II diffractometer (graphite monochromated Mo-K_αradia-

tion, λ = 0.71073 Å) by use of ω scans at –140 °C (Table 6). The

structures were solved by direct methods (SHELXS-97) and refined

on F²by using all reflections with SHELXL-97.^[81]All non-

hydrogen atoms were refined anisotropically. Most hydrogen atoms

were placed in calculated positions and assigned to an isotropic

Eur. J. Inorg. Chem. 2014, 4660–4676

4673

www.eurjic.org

FULL PAPER

Table 6. Crystallographic data, data collection and refinement details.

[(THF)_0.85(MeCN)_0.15

Cu(OH)]₂(PF₆)₂

(

^tBu,HBOX-Me₂)-

[(H₂O) (^tBu,HBOX-Me₂)-

Cu(OH)]₂(PF₆)₂

Cu^ItBu

Empirical formula

Formula weight

Crystal size [mm³]

Crystal system

Space group

a [Å]

C₁₇H₂₉CuF₆N₃O₂P

515.94

0.34ϫ0.25ϫ0.14

monoclinic

C2/c

15.8795(5)

23.5320(6)

13.7259(4)

115.655(2)

4623.4(2)

8

C_37.40H_66.50Cu₂F₁₂N_4.30O_7.70P₂

1116.67

0.34ϫ0.14ϫ0.10

monoclinic

P2₁/c

C₃₀H₅₆Cu₂F₁₂N₄O₈P₂

1017.81

0.27ϫ0.19ϫ0.11

monoclinic

P2₁/n

10.3550(4)

19.2516(9)

11.2821(5)

107.678(3)

2142.88(16)

2

9.1739(6)

b [Å]

c [Å]

12.6410(6)

20.4258(13)

95.710(5)

2357.0(2)

2

β [°]

V [Å³]

Z

ρ [g/cm³]

1.482

1.573

1.577

F(000)

2128

1.079

0.6659/0.8790

1.67–26.75

Ϯ20, Ϯ29, Ϯ17

27955

1157

1.070

0.7106/0.8113

1.90–25.67

Ϯ11, Ϯ15, Ϯ24

28616

1048

1.168

0.7182/0.9103

2.12–26.89

–13 – 12, Ϯ24, Ϯ14

27374

μ [mm^–1]

T_min/T_max

θ range [°]

hkl range

Measured refl.

Unique refl. [R_int

]

4898 [0.0571]

4133

4898/164/386

1.022

0.0355, 0.0882

0.0457, 0.0925

–0.281/0.437

4462 [0.0440]

3780

4462/76/347

1.039

0.0412, 0.1085

0.0517, 0.1135

–0.347/0.628

4572 [0.0456]

3826

4572/2/275

1.062

0.0452, 0.1021

0.0590, 0.1081

–0.569/0.740

Observed refl. [IϾ2σ(I)]

Data/restraints/param.

Goodness-of-fit (F²)

R1 [IϾ2σ(I)]

wR2 (all data)

Resid. el. dens. [e/Å³]

displacement parameter of 1.2/1.5 U_eq(C). The oxygen-bound

hydrogen atoms in [(H₂O)(^tBu,HBOX-Me₂)Cu(OH)]₂(PF₆)₂were re-

fined with a fixed isotropic displacement parameter of 0.08 Å²and

DFIX restraints for the O–H distances of 0.82 Å. Face-indexed ab-

sorption corrections were performed numerically with the program

eter hybrid functional with the correlation functional of Lee, Yang,

and Parr (B3LYP).^[85,86]The valence triple-ζ plus polarisation basis

sets (def2-TZVPP)^[87,88]were used on copper, oxygen and nitrogen

atoms and double-ζ split valence polarisation sets (def2-SVP)^[87,89]

on carbon and hydrogen, as implemented in ORCA. The resolution

X-RED.^[82]In Cu^ItBu and [(THF)_0.85(MeCN)_0.15

(

^tBu,HBOX-Me₂)- of

identity/chain

of

spheres

exchange

approximation

Cu(OH)]₂(PF₆)₂the PF₆counterions were found to be disordered

as well as parts of the ligand in case of the former and for the

coordinating solvent molecules in case of the latter compound. The

disorder of the two different solvent molecules in the latter case is

related to the disorder of the counterion. After the initial refine-

ment, the occupancy factors were fixed at 0.85 and 0.15, which

represent the fraction of the two solvents (THF and MeCN) and

(RIJCOSX),^[90–92]together with the matching def2-TZVPP/J and

def2-SVP/J auxiliary basis set were used to accelerate the calcula-

tions. Default convergence criteria were used for geometry optimis-

ation calculations (OPT); for SCF calculations, strict convergence

criteria were used (TightSCF). Slow convergence was expected

(SlowConv). Solvent effects were considered in all calculations by

invoking the conductor-like screening model (COSMO) with the

selection of THF as a solvent. A DFT spin-unrestricted formalism

(UKS) has been applied to first calculate a high-spin triplet state (S

= 1) geometry, which was then used to calculate the spin-polarised

broken-symmetry (BS) solution.^[55,56]Therefore, the spin on one

copper atom was flipped by invoking the “FlipSpin” mechanism in

ORCA in expectation of an antiferromagnetically coupled state,

M_S= 0 (FinalMs 0). Time-dependent DFT (TD-DFT) calculations

and natural transition orbital (NTO) analysis were undertaken with

ORCA version 3.0.1. Calculations on the geometry-optimised coor-

dinates were performed using the B3LYP functional together with

basis sets and COSMO model stated above. The first 16 excited

states were calculated; the maximum dimension of the expansion

space in the Davidson procedure (MaxDim) was 320. Absorption

spectra were created by the utility program ‘orca_mapspc’ and plot-

ted with the program package OriginLab OriginPro 8.5.0G. Molec-

ular structures, orbitals and spin densities were visualised with the

UCSF Chimera package.^[93]

–

the occupancy of the PF₆with respect to two alternate positions.

EADP constraints and DELU, SIMU and BUMP restraints were

applied to model the disorder. Two counterions (occupancy 0.5) in

Cu^ItBu were found to be disordered about twofold axes. Addition-

ally parts of the ligands are disordered [O2, C8, C9, C10, C11;

occupancy factors: 0.584(11)/0.416(11)]. SADI, SIMU and DELU

restraints were used to model the disorder.

CCDC-978251

(

(Cu^ItBu),

-978252

{[(THF)_0.85(MeCN)_0.15-

^tBu,HBOX-Me₂)Cu(OH)]₂(PF₆)₂} and -978253 {[(H₂O)(^tBu,HBOX-

Me₂)Cu(OH)]₂(PF₆)₂} contain the supplementary crystallographic

data for this paper. These data can be obtained free of charge from

The Cambridge Crystallographic Data Centre via http://

www.ccdc.cam.ac.uk/data_request/cif.

Theoretical Calculations: Density functional theory (DFT) calcula-

tions were performed with the ORCA 2.9.1 electronic structure pro-

gram package by F. Neese.^[83]The geometry of [{(^tBu,HBOX-Me₂)-

Cu}₂(O₂)(MeCN)₂]²⁺was fully optimised starting from coordinates

obtained from the X-ray crystal structure of [(MeCN)₂{(^tBu,HBOX-

Me₂)Cu}₂(OH)₂]²⁺after manipulation of the Cu₂O₂distances

(using the molecule editor Avogadro 1.1.0^[84]). Ligand geometries

were used without truncation. Geometry optimisations and single-

point calculations were performed by using Becke’s three-param-

Supporting Information (see footnote on the first page of this arti-

cle): Details of electrochemical experiments, DFT calculations and

kinetic measurements. Further vibrational spectra, UV/Vis spectra

and traces, Beer’s law plot, HRMS (ESI) spectra of Cu₂(OH)₂and

van-der-Waals plots of X-ray structures.

Eur. J. Inorg. Chem. 2014, 4660–4676

4674

www.eurjic.org

FULL PAPER

[28] M. Gómez, G. Muller, M. Rocamora, Coord. Chem. Rev. 1999,

193–195, 769–835.

Acknowledgments

[29] K. A. Jørgensen, M. Johannsen, S. Yao, H. Audrain, J. Thor-

hauge, Acc. Chem. Res. 1999, 32, 605–613.

[30] J. S. Johnson, D. A. Evans, Acc. Chem. Res. 2000, 33, 325–335.

[31] H. A. McManus, P. J. Guiry, Chem. Rev. 2004, 104, 4151–4202.

[32] G. Desimoni, G. Faita, K. A. Jørgensen, Chem. Rev. 2011, 111,

PR284–PR437.

This work has been carried out in the framework of the Inter-

national PhD program CaSuS (see www.casus.uni-goettingen.de)

and the European COST program CM 1003. The Karlsruhe syn-

chrotron ANKA (Germany) and Dr. Stefan Mangold are acknowl-

edged for providing beamtime and help for the XAS measurements.

[33] R. Rasappan, D. Laventine, O. Reiser, Coord. Chem. Rev. 2008,

252, 702–714.

[34] S. Dagorne, S. Bellemin-Laponnaz, A. Maisse-François, Eur. J.

Inorg. Chem. 2007, 913–925.

[1] L. M. Mirica, X. Ottenwaelder, T. D. P. Stack, Chem. Rev.

2004, 104, 1013–1045.

[35] A. Walli, S. Dechert, F. Meyer, Eur. J. Org. Chem. 2013, 7044–

[2] E. A. Lewis, W. B. Tolman, Chem. Rev. 2004, 104, 1047–1076.

[3] S. Itoh, Curr. Opin. Chem. Biol. 2006, 10, 115–122.

[4] E. I. Solomon, D. E. Heppner, E. M. Johnston, J. W. Ginsbach,

J. Cirera, M. Qayyum, M. T. Kieber-Emmons, C. H. Kjaer-

gaard, R. G. Hadt, L. Tian, Chem. Rev. 2014, 114, 3659–3853.

[5] K. A. Magnus, H. Ton-That, J. E. Carpenter, Chem. Rev. 1994,

94, 727–735.

7049.

[36] M. J. Baldwin, D. E. Root, J. E. Pate, K. Fujisawa, N. Kitajima,

E. I. Solomon, J. Am. Chem. Soc. 1992, 114, 10421–10431.

[37] L. M. Mirica, M. Vance, D. J. Rudd, B. Hedman, K. O. Hodg-

son, E. I. Solomon, T. D. P. Stack, J. Am. Chem. Soc. 2002,

124, 9332–9333.

[38] L. M. Mirica, M. Vance, D. J. Rudd, B. Hedman, K. O. Hodg-

son, E. I. Solomon, T. D. P. Stack, Science 2005, 308, 1890–

1892.

[39] L. M. Mirica, D. J. Rudd, M. A. Vance, E. I. Solomon, K. O.

Hodgson, B. Hedman, T. D. P. Stack, J. Am. Chem. Soc. 2006,

128, 2654–2665.

[40] K. E. Dalle, T. Gruene, S. Dechert, S. Demeshko, F. Meyer, J.

Am. Chem. Soc. 2014, 136, 7428–7434.

[41] M. J. Henson, V. Mahadevan, T. D. P. Stack, E. I. Solomon,

Inorg. Chem. 2001, 40, 5068–5069.

[42] K. Connors, Chemical Kinetics: The Study of Reaction Rates,

in: Solution, VCH Publishers, New York, Weinheim, Germany,

Cambridge, UK, 1990.

[43] S. V. Kryatov, E. V. Rybak-Akimova, S. Schindler, Chem. Rev.

2005, 105, 2175–2226.

[44] K. D. Karlin, S. Kaderli, A. D. Zuberbühler, Acc. Chem. Res.

1997, 30, 139–147.

[45] E. V. Rybak-Akimova, in: Phys. Inorg. Chem. React. Process.

Appl. (Ed.: A. Bakac), John Wiley & Sons, 2011, p. 109–188.

[46] E. V. Rybak-Akimova, W. Otto, P. Deardorf, R. Roesner, D. H.

Busch, Inorg. Chem. 1997, 36, 2746–2753.

[47] E. V. Rybak-Akimova, K. Marek, M. Masarwa, D. H. Busch,

Inorg. Chim. Acta 1998, 270, 151–161.

[48] N. W. Aboelella, S. V. Kryatov, B. F. Gherman, W. W. Brennes-

sel, V. G. Young Jr., R. Sarangi, E. V. Rybak-Akimova, K. O.

Hodgson, B. Hedman, E. I. Solomon, C. J. Cramer, W. B. Tol-

man, J. Am. Chem. Soc. 2004, 126, 16896–16911.

[6] K. E. van Holde, K. I. Miller, H. Decker, J. Biol. Chem. 2001,

276, 15563–15566.

[7] Y. Matoba, T. Kumagai, A. Yamamoto, H. Yoshitsu, M. Sugi-

yama, J. Biol. Chem. 2006, 281, 8981–8990.

[8] H. Decker, T. Schweikardt, F. Tuczek, Angew. Chem. Int. Ed.

2006, 45, 4546–50; Angew. Chem. 2006, 118, 4658.

[9] C. Gerdemann, C. Eicken, B. Krebs, Acc. Chem. Res. 2002, 35,

183–91.

[10] E. I. Solomon, P. Chen, M. Metz, S.-K. Lee, A. E. Palmer, An-

gew. Chem. Int. Ed. 2001, 40, 4570–4590; Angew. Chem. 2001,

113, 4702.

[11] H. Decker, in: Encycl. Inorg. Chem (Eds.: R. B. King, R. H.

Crabtree, C. M. Lukehart, D. A. Atwood, R. A. Scott), John

Wiley & Sons, Chichester, UK, 2006.

[12] N. Kitajima, K. Fujisawa, C. Fujimoto, Y. Moro-oka, S. Hash-

imoto, T. Kitagawa, K. Toriumi, K. Tatsumi, A. Nakamura, J.

Am. Chem. Soc. 1992, 114, 1277–1291.

[13] N. Kitajima, Y. Moro-oka, Chem. Rev. 1994, 94, 737–757.

[14] J. A. Halfen, S. Mahapatra, E. C. Wilkinson, S. Kaderli, V. G.

Young, L. Que, A. D. Zuberbühler, W. B. Tolman, Science

1996, 271, 1397–1400.

[15] S. Mahapatra, J. Halfen, E. Wilkinson, G. Pan, X. Wang, V.

Young Jr., C. Cramer, L. Que Jr., W. B. Tolman, J. Am. Chem.

Soc. 1996, 118, 11555–11574.

[16] R. R. Jacobson, Z. Tyeklar, A. Farooq, K. D. Karlin, S. Liu, J.

Zubieta, J. Am. Chem. Soc. 1988, 110, 3690–3692.

[17] G. Y. Park, M. F. Qayyum, J. Woertink, K. O. Hodgson, B.

Hedman, A. A. Narducci Sarjeant, E. I. Solomon, K. D. Kar-

lin, J. Am. Chem. Soc. 2012, 134, 8513–8524.

[18] Z. Hu, R. D. Williams, D. Tran, T. G. Spiro, S. M. Gorun, J.

Am. Chem. Soc. 2000, 122, 3556–3557.

[19] M. Kodera, K. Katayama, Y. Tachi, K. Kano, S. Hirota, S.

Fujinami, M. Suzuki, J. Am. Chem. Soc. 1999, 121, 11006–

11007.

[20] E. I. Solomon, J. W. Ginsbach, D. E. Heppner, M. T. Kieber-

Emmons, C. H. Kjaergaard, P. J. Smeets, L. Tian, J. S. Woert-

ink, J. Chem. Soc. Faraday Trans. 2011, 148, 11.

[21] E. I. Solomon, Inorg. Chem. 2006, 45, 8012–25.

[22] R. H. Holm, P. Kennepohl, E. I. Solomon, Chem. Rev. 1996,

96, 2239–2314.

[23] M. Rolff, J. Schottenheim, H. Decker, F. Tuczek, Chem. Soc.

Rev. 2011, 40, 4077–98.

[24] I. A. Koval, P. Gamez, C. Belle, K. Selmeczi, J. Reedijk, Chem.

Soc. Rev. 2006, 35, 814–840.

[25] B. Hazes, K. A. Magnus, C. Bonaventura, J. Bonaventura, Z.

Dauter, K. H. Kalk, W. G. Hol, Prot. Sci. 1993, 2, 597–619.

[26] K. A. Magnus, B. Hazes, H. Ton-That, C. Bonaventura, J. Bon-

aventura, W. G. Hol, Proteins: Struct., Funct., Bioinf. 1994, 19,

302–309.

[27] A. K. Ghosh, P. Mathivanan, J. Cappiello, Tetrahedron: Asym-

metry 1998, 9, 1–45.

[49]

S. V. Kryatov, E. V. Rybak-Akimova, V. L. MacMurdo, L. Que,

Inorg. Chem. 2001, 40, 2220–2228.

[50]

A. Kunishita, M. Z. Ertem, Y. Okubo, T. Tano, H. Sugimoto,

K. Ohkubo, N. Fujieda, S. Fukuzumi, C. J. Cramer, S. Itoh,

Inorg. Chem. 2012, 51, 9465–9480.

[51]

[52]

R. Díaz-Torres, S. Alvarez, J. Chem. Soc., Dalton Trans. 2011,

40, 10742–50.

J. L. DuBois, P. Mukherjee, A. M. Collier, J. M. Mayer, E. I.

Solomon, B. Hedman, T. D. P. Stack, K. O. Hodgson, J. Am.

Chem. Soc. 1997, 119, 8578–8579.

[53]

[54]

J. L. DuBois, P. Mukherjee, T. D. P. Stack, B. Hedman, E. I.

Solomon, K. O. Hodgson, J. Am. Chem. Soc. 2000, 122, 5775–

5787.

L. S. Kau, D. J. Spira-Solomon, J. E. Penner-Hahn, K. O.

Hodgson, E. I. Solomon, J. Am. Chem. Soc. 1987, 109, 6433–

6442.

[55]

[56]

A. P. Ginsberg, J. Am. Chem. Soc. 1980, 102, 111–117.

L. Noodleman, C. Y. Peng, D. A. Case, J.-M. Mouesca, Coord.

Chem. Rev. 1995, 144, 199–244.

[57]

[58]

E. Pidcock, H. V. Obias, M. Abe, H. Liang, K. D. Karlin, E. I.

Solomon, J. Am. Chem. Soc. 1999, 121, 1299–1308.

E. I. Solomon, D. M. Dooley, R.-H. Wang, H. B. Gray, M.

Cerdonio, F. Mogno, G. L. Romani, J. Am. Chem. Soc. 1976,

98, 1029–1031.

Eur. J. Inorg. Chem. 2014, 4660–4676

4675

www.eurjic.org

FULL PAPER

[59] K. D. Karlin, Z. Tyeklár, A. Farooq, R. R. Jacobson, E. Sinn,

D. W. Lee, J. E. Bradshaw, L. J. Wilson, Inorg. Chim. Acta 1991,

182, 1–3.

[60] K. D. Karlin, Z. Tyeklár, A. Farooq, M. S. Haka, P. Ghosh,

R. W. Cruse, Y. Gultneh, J. C. Hayes, P. J. Toscano, J. Zubieta,

Inorg. Chem. 1992, 31, 1436–1451.

[61] Y. Takano, K. Koizumi, H. Nakamura, Inorg. Chim. Acta 2009,

362, 4578–4584.

[62] S. Herres-Pawlis, P. Verma, R. Haase, P. Kang, C. T. Lyons,

E. C. Wasinger, U. Flörke, G. Henkel, T. D. P. Stack, J. Am.

Chem. Soc. 2009, 131, 1154–69.

[63] A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn, G. C. Ver-

schoor, J. Chem. Soc., Dalton Trans. 1984, 1349–1356.

[64] S. Itoh, M. Taki, H. Nakao, P. L. Holland, W. B. Tolman, L.

Que Jr., S. Fukuzumi, Angew. Chem. Int. Ed. 2000, 39, 398–

400; Angew. Chem. 2000, 112, 409.

[65] S. Yokota, Y. Tachi, S. Itoh, Inorg. Chem. 2002, 41, 1342–1344.

[66] D. R. Burfield, R. H. Smithers, J. Org. Chem. 1978, 43, 3966–

3968.

[74] S. J. Gurman, N. Binsted, I. Ross, J. Phys. C 1984, 17, 143–

151.

[75] M. Roy, S. J. Gurman, J. Synchrotron Radiat. 1999, 6, 228.

[76] A. Kirfel, K. Eichhorn, Acta Crystallogr., Sect. A 1990, 46,

271–284.

[77] H. Yamada, Y. Soejima, X. G. Zheng, M. Kawaminami, Trans.

Mater. Res. Soc. Jpn. 2000, 25, 1199–1202.

[78] H. R. Oswald, A. Reller, H. W. Schmalle, E. Dubler, Acta Crys-

tallogr., Sect. C 1990, 46, 2279–2284.

[79] E. Stern, Phys. Rev. B 1993, 48, 9825–9827.

[80] N. Binsted, EXCURV98: CCLRC Daresbury Laboratory

Computer Program, Manual, 1998.

[81] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.

[82] X-RED, STOE & CIE GmbH, Darmstadt (Germany), 2002.

[83] F. Neese, WIREs Comput. Mol. Sci. 2012, 2, 73–78.

[84] M. D. Hanwell, D. E. Curtis, D. C. Lonie, T. Vandermeersch,

E. Zurek, G. R. Hutchison, J. Cheminf. 2012, 4, 17.

[85] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.

[86] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.

[87] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7,

3297–3305.

[67] G. J. Kubas, B. Monzyk, A. L. Crumblis, Inorg. Synth. 1979,

19, 90–92.

[68] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 887–910.

[69] J. R. Aranzaes, M.-C. Daniel, D. Astruc, Can. J. Chem. 2006,

84, 288–299.

[70] R. Battino, T. R. Rettich, T. Tominaga, J. Phys. Chem. Ref.

Data 1983, 12, 163.

[71] E. Bill, julX Simulation Program for Static Molar Magnetic

Susceptibilities, Max Planck Institute For Chemical Energy

Conversion, Mülheim an der Ruhr, Germany, eckhard. [92] S. Kossmann, F. Neese, Chem. Phys. Lett. 2009, 481, 240–243.

bill@cec.mpg.de.

[72] T. S. Ertel, H. Bertagnolli, S. Hückmann, U. Kolb, D. Peter,

Appl. Spectrosc. 1992, 46, 690–698.

[73] M. Newville, P. Livins, Y. Yacoby, J. Rehr, E. Stern, Phys. Rev.

B 1993, 47, 14126–14131.

[88] A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100,

5829–5835.

[89] A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97,

2571–2577.

[90] F. Neese, J. Comput. Chem. 2003, 24, 1740–1747.

[91] F. Neese, F. Wennmohs, A. Hansen, U. Becker, Chem. Phys.

2009, 356, 98–109.

[93] E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch,

D. M. Greenblatt, E. C. Meng, T. E. Ferrin, J. Comput. Chem.

2004, 25, 1605–1612.

Received: May 2, 2014

Published Online: August 1, 2014

Eur. J. Inorg. Chem. 2014, 4660–4676

4676

Article Doi

BOX ligands in biomimetic copper-mediated dioxygen activation: A hemocyanin model

DOI: 10.1002/ejic.201402378

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejic.201402378

Full text of DOI:10.1002/ejic.201402378

Products guided by the article

R&D Labs maybe for 6390-69-8

Relevant to this article

Hot Product