A new family of multinuclear mixed-ligand copper(II) clusters: Crystal structures, magnetic properties and catecholase-like activity

Article abstract of DOI:10.1016/j.ica.2016.07.057

In this work we report a family of mixed-ligand copper(II) clusters obtained with different nuclearities: [Cu₃(μ-OH)(μ-dppi)₃(μ-hfac)₃] (1), [Cu₄(μ-dppi)₄(Hdppi)₂(hfac)₄] (2), [Cu₅(μ-O)₂(μ-dppi)(μ-Hdppi)₂(μ-hfac)₃(hfac)₂] (3) and [Cu₆(μ-O)₂(μ-dppi)₂(μ-Hdppi)(μ-hfac)₆] (4), where hfac?=?hexafluoroacetylacetonate, dppi?=?diphenylphosphinate and Hdppi?=?diphenylphosphinic acid. Complexes 3 and 4 are the first examples of copper(II) clusters containing a μ-oxo oxygen atom and hfac as bridging ligands. The magnetic properties of all compounds showed predominant antiferromagnetic interactions and the magnetic coupling constants were evaluated using an isotropic exchange model based on the spin topology. A spin frustration phenomena was observed for the equilateral cyclic trinuclear complex (1), thus an antisymmetric exchange term was also considered to fit the experimental magnetic data. The obtained J values are in the range from ?89?cm^?1to 1.7?cm^?1and were interpreted based on structural features. The electrochemical properties and catalytic activity of complexes 1 and 2 were investigated and compared with the [Cu(hfac)₂] precursor. Catecholase-like activity of copper(II) complexes via oxidation of 3,5-di-tert-butylcatechol (3,5-DTBC) to 3,5-di-tert-butylquinone (3,5-DTBQ) was performed.

Full text of DOI:10.1016/j.ica.2016.07.057

Accepted Manuscript
A new family of multinuclear mixed-ligand copper(II) clusters: crystal struc-
tures, magnetic properties and catecholase-like activity
Sergiu Calancea, Samira G. Reis, Guilherme P. Guedes, Rafael A. Allão
Cassaro, Felipe Semaan, Fernando López-Ortiz, Maria G.F. Vaz
PII:
S0020-1693(16)30440-6
http://dx.doi.org/10.1016/j.ica.2016.07.057
ICA 17192
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 April 2016
28 June 2016
29 July 2016
Please cite this article as: S. Calancea, S.G. Reis, G.P. Guedes, R.A. Allão Cassaro, F. Semaan, F. López-Ortiz,
M.G.F. Vaz, A new family of multinuclear mixed-ligand copper(II) clusters: crystal structures, magnetic properties
and catecholase-like activity, Inorganica Chimica Acta (2016), doi: http://dx.doi.org/10.1016/j.ica.2016.07.057
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A new family of multinuclear mixed-ligand copper(II) clusters:
crystal structures, magnetic properties and catecholase-like activity
Sergiu Calancea,† Samira G. Reis,† Guilherme P. Guedes,§ Rafael A. Allão Cassaro,⊥
Felipe Semaan,† Fernando López-Ortiz,*⊗ and Maria G. F. Vaz*,
†
†Instituto de Química Universidade Federal Fluminense, Niterói, Rio de Janeiro, Brazil.
§Instituto de Ciências Exatas, Departamento de Química, Universidade Federal Rural do Rio de Janeiro,
Seropédica, Rio de Janeiro, Brazil.
⊥Instituto de Química, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil.
⊗Área de Química Orgánica, Universidad de Almería, Crta, Sacramento s/n, 04120, Almería, Spain.
* Corresponding author
E-mail address: mariavaz@vm.uff.br(Maria Vaz) and flortiz@ual.es
(Fernando L. Ortiz)
Abstract
In this work we report a family of mixed-ligand copper(II) clusters obtained with different
nuclearities:
[Cu3(µ-OH)(µ-dppi)3(µ-hfac)3]
(1),
[Cu4(µ-dppi)4(Hdppi)2(hfac)4]
(2),
[Cu5(µ-O)2(µ-dppi)(µ-Hdppi)2(µ-hfac)3(hfac)2] (3) and [Cu6(µ-O)2(µ-dppi)2(µ-Hdppi)(µ-hfac)6]
(4) complexes, where hfac = hexafluoroacetylacetonate, dppi = diphenylphosphinate and Hdppi
= diphenylphosphinic acid. Complexes 3 and 4 are the first examples of copper(II) clusters
containing a µ-oxo oxygen atom and hfac as bridging ligands. The magnetic properties of all
1

compounds showed predominant antiferromagnetic interactions and the magnetic coupling
constants were evaluated using an isotropic exchange model based on the spin topology. A spin
frustration phenomena was observed for the equilateral cyclic trinuclear complex (1), thus an
antisymmetric exchange term was also considered to fit the experimental magnetic data. The
obtained J values are in the range from -89 cm-1 to 1.7 cm-1 and were interpreted based on
structural features. The electrochemical properties and catalytic activity of complexes 1 and 2
were investigated and compared with the [Cu(hfac)2] precursor. Catecholase-like activity of
copper(II) complexes via oxidation of 3,5-di-tert-butylcatechol (3,5-DTBC) to 3,5-di-tert-
butylquinone (3,5-DTBQ) was performed.
Keywords: copper(II) complexes, mixed-ligand complexes, magnetism, phosphinate,
Catecholase-like activity
1) Introduction
Polynuclear metal complexes (PMCs) are a hot topic nowadays due to the application of
some of these compounds as catalysts,[1,4] luminescence systems,[5,6] electronic devices,[7]
semiconductors,[8,9] molecular-based magnetic[10-15] materials and mimics of the active site of
metalloenzymes.[16-18] A large number of copper(II)-based PMCs with different architectures
and nuclearities can be constructed by using this cheap and readily available metal.[1-18]
However, the synthesis of polynuclear copper(II) clusters in a controllable way is challenging.
Apart from the major role played by the ligand or combination of ligands selected in assembling
the units leading to the coordination framework, other factors such as solvent, temperature,
counter-ion and additives can be determinant for the definitive structure and properties of the
complex.[19,20] An useful approach to obtain high-nuclearity copper clusters relies on the use of
multiple functional bridging and/or chelating neutral or anionic ligands. The carboxylate group
C(O)O- is a popular component of these multinucleating agents.[21] Similar to the carboxylate
2

linkage, the ability of the phosphinate moiety P(O)O- to act as a bridging ligand have been
extensively investigated. Although this kind of ligands have been widely used to prepare
organic-inorganic hybrid polynuclear complexes of main group and transition metal ions such as
titanium, manganese, cobalt, nickel, silver, etc,[22-28] very few copper complexes built on the
exclusive basis of phosphinate ligands have been described to date. Generally, the 2.11 (II and
III) coordination mode[29, 30] predominates, i.e., µ-O,O’-bridges (Scheme 1),[31-37, 38] even
though a 3.21 (IV) binding mode has been found in a dimethylphosphinate complex. [39]
Scheme 1. Coordination modes of phosphinate in transition metal complexes (R1, R2 = alkyl,
aryl).[29, 30]
A few examples of heteroleptic copper(II) complexes containing phosphinate ligands have
been previously described.[40-42] In these cases, the phosphinate groups act as a bridging ligand
between copper ions, leading to chains and dinuclear compounds. In addition, some examples of
mononuclear[43,44] and dinuclear[38] copper(II)-diphenylphosphinate complexes in which the
ligand is bound to the metal through only one up to the two oxygen atoms are also known.
3

However, to the best of our knowledge, no copper(II) cluster assembled through the phosphinate
group has been reported until now.
Binuclear and trinuclear copper clusters are present in multicopper oxidases (MCOs). These
are a family of enzymes featuring at least four copper ions per functional unit that nature has
developed for the oxidation of their substrates with concomitant four-electron reduction of O2 to
water.[45,46] Mechanistic studies suggested the participation in this process of oxygen bridged
metal−dioxygen intermediates.[47,48] Due to the technological and economical relevance of the
four-electron reduction of O2,[49-51] intensive efforts have been devoted to the development of
simpler and more accessible synthetic multicopper systems to achieve this process.[52-55] Many
complexes containing several Cu−O2 binding modes that mimic the behavior of MOCs have
been described. Inspired by these works, we thought that the complexation of copper(II)
hexafluoroacetylacetonate with diphenylphosphinic acid could be directed to the formation of
new copper(II) coordination clusters by using dioxygen as an organizing agent of the metal
centers. Hexafluoroacetylacetonate ligand is extensively used as building block in molecular
magnetism area.[56-60] However, there are no precedent mixed-ligand copper(II) compounds
containing hfac and phosphinates.
In this contribution, we report the synthesis and crystal structure of the mixed-ligand
copper(II) clusters [Cu3(µ-OH)(µ-dppi)3(µ-hfac)3] (1), [Cu4(µ-dppi)4(Hdppi)2(hfac)4] (2),
[Cu5(µ-O)2(µ-dppi)(µ-Hdppi)2(µ-hfac)3(hfac)2] (3) and [Cu6(µ-O)2(µ-dppi)2(µ-Hdppi)(µ-hfac)6]
(4). Complexes 1-3 are obtained from the reaction of [Cu(hfac)2] with Hdppi in the presence of
dioxygen while complex 4 is formed in the reaction of [Cu(hfac)2] with dppnTEMPO radical (1-
piperidinyloxy-4-[(diphenylphosphinyl)amino]-2,2,6,6-tetramethyl) via hydrolysis of the P(O)-N
linkage. The careful selection of the reaction conditions allowed us to obtain complexes with
4

different nuclearities varying from trinuclear to hexanuclear in a reproducible way. Complexes 3
and 4 are the first examples of copper(II) clusters having a µ-oxo-independent oxygen atom with
hfac as bridging ligands. The magnetic properties of all compounds were studied, and in addition,
the electrochemical properties and catalytic activity of complexes 1 and 2 in a redox reaction
were investigated. Catecholase-like activity of copper(II) complexes via oxidation of 3,5-di-tert-
buthylcatechol (3,5-DTBC) to 3,5-di-tert-buthylquinone (3,5-DTBQ) was performed.
2) Experimental section
2.1. General procedures
All the reagents and solvents were purchased from commercial sources and directly handled
without further purification.
2.2. Synthesis of complexes 1 and 3
[Cu(hfac)2(H2O)2] (1: 117 mg, 0.23 mmol, and 3: 24 mg, 0.039 mmol) was dissolved in 50
mL of boiling n-heptane followed by continuous bubbling of oxygen gas. The solution was
concentrated until ca. 15 mL, then a chloroform solution (2 mL) of diphenylphosphinic acid (1:
50 mg, 0.23 mmol; and 3: 10 mg, 0.046 mmol) was added. For the synthesis of 1, triethylamine
(32 µL, 0.23 mmol) was added under vigorous stirring. Addition of Ph2P(O)OH caused solution
color changes from dark-green to light green-yellow and the appearance of small amounts of a
pale blue precipitate. Insoluble material was removed through filtration. After a few hours (1) or
2-3 days (3) dark-green crystals were obtained. 1: Yield 38%. Anal. Calc. for
C57H52Cu3F18NO14P3 (1·Et3NH+·H2O): C, 42.77 ; H, 3.27 ; N, 0.88. Found: C, 42.28 ; H, 3.57 ;
5

N, 1.10. IR (ATR, ν cm-1): 3058, 3013 (w) (νC-HAr), 1646 (s) (νC=O), 1545 (s) (νC=C), 1260
(s) (νP=O), 1192 (s) 1131 (s), 1033 (s) (νC-F). Complex 3: Yield 11%. Anal. Calc. for
C61H37Cu5F30O18P3: C, 35.94 ; H, 1.83. Found: C, 36.39 ; H, 1.97. IR (ATR, ν cm-1): 1644 (m)
(νC=O), 1433 (m) (νC=C), 1267 (s) (νP=O), 1107 (sb), 1011 (sb) (νC-F).
2.3. Synthesis of 2
The [Cu(hfac)2(H2O)2] (235 mg, 0.47 mmol) was dissolved in 50 mL of boiling n-heptane
followed by continuous bubbling of oxygen gas. The volume of the solution was reduced to ca.
15 mL and then a solution of 100 mg of diphenylphosphinic acid (0.46 mol) in 2 mL of methanol
was added. Bubbling of oxygen was continued during 30 min. After 2-3 days, dark green crystals
were obtained. Yield 23%. Anal. Calc. for C92H66Cu4F24O20P6: C, 46.28 ; H, 2.70. Found: C,
46.64 ; H, 2.74. IR (ATR, ν cm-1): 3062 (w) (νC-HAr), 1644 (s) (νC=O), 1459 (s) (νC=C), 1243
(s) (νP=O), 1131 (sb), 1050 (m) (νC-F).
2.4. Synthesis of 4
Complex 4 was obtained from the reaction of [Cu(hfac)2(H2O)2] with dppnTEMPO radical
through in situ hydrolysis of the P(O)-N linkage as secondary product together with the main
product already described in our previous work.[61] 26 mg (0.053 mmol) of [Cu(hfac)2(H2O)2]
was added to 10 mL of n-heptane and boiled until dissolution was complete. Then, 20 mg (0.053
mmol) of the dppnTEMPO radical was dissolved in CHCl3 and quickly added under constant
stirring.
After
a
few
days,
dark-green
crystals
of
[Cu(hfac)2(dpnnTEMPO)2][Cu(hfac)2(dpnnTEMPO)] were obtained. These crystals were filtered
0
out and the filtrated kept at 15 C. After 2 months, dark-blue crystals formed from the filtrate
solution were separated manually from the main compound. Yield 30%. Anal. Calc. for
6

C66H37Cu6F36O20P3·0.5H2O, 34.21%; H, 1.65%. Found: C, 34.66%; H, 1.84%. IR (ATR, ν cm-1):
1638 (s) (νC=O), 1452 (s) (νC=C), 1217 (s) (νP=O), 1139 (sb), 991 (mb) (νC-F).
2.5. Physical Measurements
CHN elemental analyses were performed on a Perkin Elmer 2400. Infrared spectra (4000-
400 cm-1) were recorded on a Bruker Alpha-P FTIR spectrometer. Electronic spectra were
acquired from THF solutions on a Varian 1200 spectrophotometer. The magnetic measurements
were performed on a bunch of single crystals using a Cryogenic Sx-600 SQUID in the
temperature range 2-280 K and field range of 0 to 6.5 T. The sample was placed in a gelatine
capsule and data were corrected for the sample diamagnetism and sample holder. The magnetic
data were fitted using the Magprop software, available within the DAVE package.[62]
Electrochemical studies were achieved on Metrohm Autolab potentiostat PGSTAT101 with
Pt (Ø3 mm) as working and counter electrode, and Ag/AgCl as a reference electrode. All
manipulations were carried out at 298 K in THF with 0.10 mol·L-1 tetrabutylammonium
hexafluorophosphate, Bu4NPF6, as supporting electrolyte. Sweep rate was 50-100 mV/s.
Single crystal X-ray data of 1-3 complexes were collected on a Bruker D8 Venture
diffractometer using graphite monochromatic MoKα radiation (λ = 0.71073 Å). Data collection
and cell refinement were performed with Bruker Instrument Service v4.2.2 and APEX2,[63]
respectively. Data reduction was carried out using SAINT.[64] Empirical multiscan absorption
correction using equivalent reflections was performed with the SADABS program.[65] For 4, the
crystallographic data were collected on an Oxford GEMINI A Ultra diffractometer using
graphite monochromated Mo Kα radiation. Data collection, data reduction, cell refinement and
absorption corretions were performed by using the CrysAlis RED, Oxford Diffraction Ltd.,
Version 1.171.32.38. The structures solution and refinement were performed with the SHELXS-
7

2014 and SHELXL-2014 programs.[66] All atoms except hydrogen were refined anisotropically.
The H-atoms were treated using a constrained refinement. Despite of our best efforts it was not
possible to achieve a good refinement for compound 1. It is possible to say that it crystallizes in a
trigonal crystal system and the molecular structure consists of a trimetallic triangular core, in
which the copper(II) ions are linked by bridging hfac, dppi and one µ3-hydroxo ligand (see Fig.
S1). Details of data collection and structure refinement for compounds 2-4 are summarized in
Table 1.
Table 1: Summary of crystal data, collection and refinement for compounds 2-4.
2
3
4
Chemical formula
Formula weight
Crystal system
2(C46H33Cu2F12O10P3)
1193.71
Triclinic
13.6349(17)
13.6603(16)
15.7376(19)
84.342(4)
75.042(3)
61.085(3)
2477.7(5)
293(2)
C61H37Cu5F30O18P3
2038.51
Monoclinic
15.15877 (13)
21.278 (2)
22.9840 (18)
90
C66H37Cu6F36O20P3.(H2O)0.5
2316.11
Monoclinic
22.5132(6)
14.0879(4)
27.5041(10)
90
a/Å
b/Å
c/Å
α/º
β/º
98.975 (3)
90
7322.2 (1)
293(2)
106.332(3)
90
8371.3(4)
120(2)
γ/ º
Unit cell volume/Å3
Temperature/K
Space group
P-1
P21/n
C c
Z
2
4
4
Radiation type
MoKα
1.056
MoKα
1.638
MoKα
1.702
µ/mm-1
2.1-25.7
25867
9334
2.0-27.3
65017
16368
1.9-25.0
26074
12239
θ range /º
reflections measured
independent reflections
Rint
0.056
0.070
0.049
0.075
0.192
0.120
0.213
0.047
0.104
0.083
0.1213
0.0587
0.1418
0.0758
0.1560
R1 values (I>2σ(I))
wR(F²) values (I>2σ(I))
R1 values (all data)
wR(F²) values (all data)
Goodness of fit on F²
Largest diff. peak and role (e. Å-3)
1.072
1.020
1.268; -0.976
1.036
2.086; -0.640
0.888;-0.536
2.6. Catalytic oxidation of 3,5-DTBC
Catecholase-like activity was determined using 10-4 mol·L-1 solutions of each copper
complex in THF. In order to keep the same number of CuII ions in solution, the molar
8

concentration of complexes 1, [Cu(hfac)2] and 2 was calculated dividing their molecular weights
by three (Cu3), one (Cu) and four (Cu4), respectively. The UV spectra were recorded in 1 min
interval following the increasing of absorption values at λ = 392 nm during 20 min at 25 °C. All
experiments were carried out using air and dioxygen-saturated THF solutions. To determine the
dependence of the rates and kinetic parameters such as Vmax, Km and kcat, the kinetics
measurements were performed using 3,5-DTBC in the concentration range from 5ꢀ10-2 to 8ꢀ10-4
mol·L-1.
3) Results and discussion
3.1. Syntheses
Complexes 1-3 have been prepared according to the synthetic routes shown in Scheme 2a.
The complex 1 was obtained by reacting diphenylphosphinic acid and [Cu(hfac)2] in 1:1
stoichiometry in a n-heptane-chloroform solution, using triethylamine as a base and under
continuously bubbling of dioxygen gas. The analogous reaction of [Cu(hfac)2] with
diphenylphosphinic using n-heptane-methanol solution provided the complex 2. Increasing to
two equivalents the amount of [Cu(hfac)2] that reacted with diphenylphosphinic acid (1:2
stoichiometry) in a n-heptane-chloroform solution lead to the formation of the complex 3.
Complex 4 proceeds from the reaction of [Cu(hfac)2] with dppnTEMPO radical using a
stoichiometric ratio [Cu(hfac)2]:radical of 1:1 in a n-heptane-chloroform solution (Scheme 2b).
Formation of 4 involved in situ hydrolysis of the P-N bond of the radical to give
diphenylphosphinic acid that acts as bridging ligand towards copper(II) ions. Crystals of
complex 4 were isolated after growing for 2 months in the reaction medium and were separated
9

manually
from
the
main
compound
identified
as
the
complex
[Cu(hfac)2(dpnnTEMPO)2][Cu(hfac)2(dpnnTEMPO)] previously reported.[61]
Scheme 2. Synthesis of the complexes 1-3 (a) and 4 (b)
3.2. Structure descriptions
The coordination modes of dppi in complexes 1-4 include binding types I, II, III shown in
Scheme 1, and a new one that has not been previously reported, syn-3.21 (VI, Scheme 3).[29, 30]
As for the hfac ligand, two coordination modes are present, the usual chelate syn-1.11 (VII) and
the first example of a homonuclear copper(II) complex showing a syn-2.11 (VIII) coordination
(Scheme 3).
10

VI
VII
VIII
Scheme 3. Coordination modes of the dppi and hfac ligands in complexes 1-4.
Despite extensive experimentation, good quality single crystals of complex 1 could not be
obtained. Notwithstanding, the analysis of the X-ray data collected allowed to identify the
molecular structure of this compound as a trimetallic equatorial triangular core, in which the
copper(II) ions are joined together by three hfac, three dppi ligands and one independent hydroxo
ligand (Fig. S1).
Compound 2 crystallizes in triclinic P-1 and the molecular structure consists of a
tetranuclear copper(II) cluster, in which metal ions are held together by four dppi, two Hdppi and
four hfac ligands as seen in Fig. 1. The four copper(II) centers are pentacoordinated, and in this
geometry the metal environment can be described between square-pyramidal (SP) and
trigonal-bipyramidal (TBP) extremes. In terms of the parameter τ, these extremes are
represented by the values τ = 0 and τ = 1 for the SP and TBP geometries, respectively.[67]
The calculated τ = 0.16 indicates a distorted square pyramidal geometry with equatorial
positions occupied by two hfac oxygen atoms [O3, O4 for Cu1 and O1, O2 for Cu2], one dppi
(or Hdppi) oxygen atom [O2, O5 for Cu1; O1, O7 for Cu2] and in apical position by one dppi (or
Hdppi) oxygen atom [O7 for Cu1 and O10 for Cu2]. All hfac ligands are coordinated in a
chelating 1.11 mode (VII), while the dppi ligands are bridged coordinated in the 3.21 (VI) and
2.11 (II) modes (Scheme 2 and 3). The terminal Hdppi ligands are coordinated to Cu1 ions in the
1.10 mode (I). The coordination modes of dppi and Hdppi with respect to the Cu4 plane are
11

shown in Figure S2. Selected bond lengths and bond angles are listed in Table S1. The
assignment of the protonated Hdppi or dppi ligands in the structure was made by a careful
analysis on the P-O bond distances, which ranges between 1.501(5) and 1.522(6) Å. It is
important to stress that the diphenylphosphinate group in compounds 2-4 is bridged coordinated
to two or more metal ions in asymmetric and symmetric modes, while the diphenylphosphinic
acid is coordinated also in monodentated mode through the P=O moiety. The shortest P-O bond
length was attributed to the phosphinyl group (P=O), as also described for the complex
[Cu(Hdppi)(bdmpp)]2(dppi)2, where Hbdmpp = 1,3-bis(3,5-dimethylpyrazol-1-yl)-propan-2-
ol[38]. The equatorial Cu-Odppi bond lengths range from 1.887(5) to 1.971(6) Å, while the axial
Cu-Odppi are 2.712(5) Å for Cu1-O7 and 2.634(5) Å for Cu2-O10i (i=-x,-y,-z). These bond
lengths are in agreement with other reported phosphinic copper(II) based complexes[36, 38-39].
The axial Cu1-O7dppi bond length (2.712(5) Å) is among the longest reported for copper(II)
complexes.[38] This long bond length can be attributed to steric hindrance around the metal ion
associated to Jahn-Teller axial elongation.[68,69] The intermolecular O10 to O10i distance
within the tetranuclear core of 2.414(8) Å is significantly longer than the O-O bonds found in
peroxydiphosphate compounds [P2O8]4- (1.49 Å). [70] Based on the charge balance as well as
coordination modes of Hdppi/dppi, we conclude that the bridging central ligands are
deprotonated. The four copper(II) ions of the complex define a planar parallelogram with the two
sides being the shortest intramolecular distance of 3.577(1) [Cu(1)…Cu(2) or Cu1i…Cu2i, i =-x,-
y,-z] and 5.653(1) (Cu2…Cu2i). These distances are close to those seen in the dimer
[Cu(HdppiH)(bdmpp)]2(dppi)2[38] and in the chain of bis(dibutylphosphinato)copper(II),[33] in
which the metal ions are bridged by dppi ligand in the 3.21 (IV) and 2.11 (II) modes. Other
intramolecular copper-copper distances are 7.685(2) Å [Cu1…Cu2i or Cu1i…Cu2] and 10.571(2)
12

Å (Cu1…Cu1i). All dppi ligands are in trans positions of the plane of the Cu4 core (Figure S2). If
the dppi ligands in 2.11 (II) coordination mode bind the copper(II) ions outside of the Cu4 plane,
then the 3.21 (VI) and 1.10 (I) ones are closer to the metallic core plane. Hydrogen-bonding
involving terminal Hdppi ligands are observed between the Cu4 molecular units, leading to a
supramolecular 1D chain along the c axis. The shortest intermolecular distance between the
copper(II) ions is 5.851(1) Å [Cu1…Cu1i].
Complex
and bond angles for
of complex consists of a discrete pentanuclear copper(II) cluster. In this compound, the
3
crystallizes in the monoclinic P21/n space group. Selected bond lengths
3
are listed in Table S2. As shown in Figure 1, the asymmetric unit
3
copper ions are held together by one dppi, two Hdppi, five hfac and two independent oxo
ligands. Copper ions Cu1, Cu2 and Cu3 are six-coordinated with distorted octahedral
geometries, whereas the metal ions Cu4 and Cu5 are pentacoordinated with distorted
square-pyramidal geometry (τ = 0.34 for Cu4 and τ = 0.21 for Cu5). Cu1, Cu2 and Cu3 ions
are connected by two hfac ligands in the unique 2.11 mode (VIII), being O10 and O13
the bridging atoms, two 3.21 (VI) Hdppi and one independent µ3-oxo ligand O1. The
Hdppi and dppi were assigned based on the difference observed for P-O bond length. In
contrast, Cu4 and Cu5 ions are linked together by one dppi (O5 and O6) and a µ2-oxo
ligand (O2). The moieties Cu1-Cu3 and Cu4-Cu5 are connected to each other by two
Hdppi (O4 and O7) and one dppi (O5 O6) ligands in the coordination mode already
described for
3. Due to the Jahn-Teller distortion all metal ions have short Cu-O
equatorial bond lengths in the range 1.922(3)-1.985(2) Å and long axial ones [2.654(3)-
2.716(3) Å for Cu1, Cu2 and Cu3 ions; 2.375(2) and 2.687(3) Å for Cu4 and Cu5,
respectively]. The average Cu-O-Cu bond angles varies depending on chelate ligand
13

being for oxo bridge, hfac and dppi and Hdppi ligands in the range 111.1(1)-130.4(1)°,
86.9(1)-89.1(1)° and 92.21-98.4(1)°, 97.5(1)-105.4(1)º, respectively. The Cu···Cu
distances in the plane formed by four copper ions [Cu2-Cu5] are in the range 3.2728(6)-
3.5158(6) Å. The Cu1 ion is above the plane containing the other four copper(II) ions
with the intramolecular distances 3.2265(6), 3.2935(6), 4.9048(6) and 4.8137(6) Å
between Cu1 and the respective ions Cu2, Cu3, Cu4 and Cu5. This complex shows short
intermolecular F···F interactions that contribute to stabilize the crystal packing (F3···F20 =
2.805(4) Å,
107.1(3)°; F5···F19i = 2.932(5)Å,
C51-F18···F23Ai = 107.1(3)°, ii = 1/2+x, 1/2-y,1/2+z).
∠
C17-F3···F20 = 157.1(3)°; F4···F19i = 2.794(5) Å;
∠
C13-F4···F19i =
∠
C13-F5···F19i = 100.4(2), i = -x, -y, -z; F18···F23Aii
= 2.77(1) Å;
∠
Compound 4 crystallizes in the monoclinic Cc space group and the molecular unit is
shown in Figure 1. Selected bond lengths and bond angles are displayed in Table S3. The
molecular structure consists of a discrete hexanuclear copper cluster with one crystallization
water molecule. In this compound, the copper(II) ions are held together by two dppi, one Hdppi,
six hfac ligands and two oxo-oxygen atoms (Figure 1). The hexanuclear core is formed by two
Cu3 triangular units connected to each other by the dppi and Hdppi ligands. For Hdppi and dppi
ligands two are bridging in 4.22 mode (V) and one in 3.21 (VI) coordination mode. All hfac
ligands are involved in 2.11 (VIII) coordination mode, in contrast with compound 3 in which
some of the hfac coordinates in the 1.11 (VII) mode (Scheme 3). Five copper ions (Cu1, Cu3,
Cu4-Cu6) have a distorted octahedral geometry being coordinated to two dppi oxygen atoms,
three oxygen atoms provided by two hfac ligands and one independent oxo ligand. The sixth
copper ion (Cu2) is pentacoordinated linked by a bridging dppi ligand through an oxygen atom
only. The calculated τ parameter is 0.16, indicating a slightly distorted SP coordination
14

environment. Due to the Jahn-Teller distortion, the axial Cu-O bond lengths of six-coordinated
copper ions are longer [2.531(4) - 2.788(7) Å] than equatorial ones [1.910(6) - 2.02(2) Å]. The
same effect is observed for Cu2 ion. The apical Cu-O length in the square pyramid of 2.712(8) Å
is significantly longer than average Cu-O bond length of 1.941(6) Å in its basal plane. The
average Cu…Cu distances in the triangular units formed by Cu1 to Cu3 and Cu4 to Cu6 metal
ions are 3.207(6) Å (average), while the distance between two opposite copper ions of distinct
triangular units is 3.500(1) Å (average). The Cu-O-Cu angles varies depending on the bridging
ligands being for oxo bridge, hfac, and dppi ligands in the range 111.2(3)-114.7(3)°, 83.0(2)-
86.9(3)° and 94.4(3)-101.5(2)°, respectively. Short contacts Csp2-H···F, Csp2-H···Csp2 and F···F
contribute to stabilize the crystal packing (F4···F12A = 2.78(2) Å, ∠C6A-F12A···F4i = 115(1)°; i
= x, -y, 1/2 + z; F13···F32 = 2.83(2) Å, ∠C26-F32···F13 = 131(1)°).
15

(b)
(a)
(c)
Figure 1. Molecular structures of complex 2 (a), 3 (b) and 4 (c). Crystallization solvent
molecules, fluorine and hydrogen atoms were omitted for the sake of clarity. The dppi and Hdppi
carbon atoms are represented in green while hfac ligands are represented in blue.
3.3. Magnetic Studies
The temperature dependence of the magnetic susceptibility of complex
1, represented as
χT, is shown in Figure 2. At 270 K, χ
T is 1.16 cm3·mol-1·K, smaller than the expected value
16

(1.24 cm3·mol-1·K) for three non-interacting copper(II) ions with
g = 2.1. Upon cooling, χT
decreases gradually until 140 K, followed by faster decrease down to 0.21 cm3·mol-1·K at
2.0 K. This behavior indicates the presence of dominant antiferromagnetic interaction. Since
the three copper(II) ions are equivalent by symmetry, the magnetic interaction was initially
estimated by fitting
χ
T considering equilateral triangle with an isotropic exchange (Hexch.
the conventional Heisenberg-Dirac-van Vleck Hamiltonian
where is the magnetic exchange interaction. An
) was considered using the mean-field approximation to
)
described
by
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
H = −J(S1 • S2 + S2 • S3 + S1 • S3 )
J
intermolecular interaction (z
J
describe the data at low temperatures, and the best fit curve was found for
g = 2.12, J1 = -16
cm-1 and z = -1.9 cm-1 (see Figure S3 for details). The shortest intermolecular distances
J
between adjacent trinuclear units are large (
interaction obtained is too large and cannot be considered as realistic. It is known that cyclic
trinuclear copper(II) complexes can show geometric spin frustration.[71,72] In fact, the
≈ 9 Å), and the intermolecular magnetic
χT
value at the lowest temperature is smaller than expected for a magnetically isolated
copper(II) ion (0.4 cm3 mol-1
·K). In this case, to interpret the magnetic data would be
·
necessary to go further than the isotropic exchange model and consider also an
antisymmetric exchange interaction[73,74] [Gij(Si×Sj)] according to the Hamiltonian
described in Equation 1, where Gij is the antisymmetric exchange vector,
∑
β
is the Bohr
Magneton and g//
,
g┴ are the parallel and perpendicular component of the g value. In an
z
equilateral triangle (D3h symmetry) Gx
=
Gy =0 and all pairs of Gij are equal (G12
=
G23
=
G31). The effective antisymmetric exchange constant,
G, is related to the sum of the
components of two interacting centers, Gij, through the expression
G
= (Gz/6)
√3.The best fit
parameters were found for:
J
= -20 cm-1, g// = 2.30, g┴ (fixed)=2.00 and
G
= 3 cm-1 with R =
17

2
3.5x10-2, where R = ∑[
(TIP) of 101x10-6 cm3
(
χTexp.
-
χ
Tteor.)2/
χ
Texp ]. A Temperature Independent Paramagnetism
·
mol-1 was considered. The
J and G values found for 1 are smaller than
previously reported for other trinuclear copper(II) complexes[75] since those complexes
have different bridging ligands in addition to µ3-OH. Although, it was found some variance
in the
J
and
G
values, the ratio |
G/
J| ~ 0.15 found for
1
is close to the value (|G/J| ~ 0.17-
0.19) reported for those complexes.[75]
H = H exch. + H ASE +H zeeman
(Eq. 1)
H exch = − J1 (S 1.S 2 + S 1.S 3 + S 2 .S 3 )
H ASE = Gz [S1 × S 2 + S 2 × S 3 + S 3 × S1 ]
{
}
Hzeeman = β g⊥[Hx (S1x + S2x + S3x ) + H y (S1y + S2 y + S3 y )]+ g// Hz (S1z + S2z + S3z )
Figure 2. Temperature dependence of χT for 1. The black line represents a fit using the
parameters described in the text (Eq. 1).
18

The temperature dependence of the magnetic susceptibility of 2, represented as χT is
shown in Figure 3. In the temperature range 60-235 K,
cm3·mol-1·K which is lower than the expected one (1.65 cm3
T decreases down to 1.23 cm3·mol-1·K at 2.5 K.
χT has a constant value of 1.58
·
mol-1
·K) for four non interacting
copper(II) ions with
g = 2.10. Below 60 K, χ
This behavior suggests the presence of dominant antiferromagnetic interaction. To estimate
the magnetic interaction, T curve was fitted assuming the spin topology outlined in Figure
using the Hamiltonian described in Eq. 2. The spin topology model for was built
χ
4
2
considering the shortest distances between copper(II) ions bridged by phosphinate groups
that act as magnetic exchange pathways. The magnetic interaction between Cu1 and Cu2 is
represented by J1 while the magnetic interaction between Cu2...Cu2i is represented by J2
.
Since the distance between two Cu1 ions represented by S1 and S3 as well as the distances
Cu1...Cu2i (-
x, -y, -z) represented by S1, S4 and S2, S3 are large, the magnetic interactions
between them are expected to be weaker and were neglected to prevent overparametrization.
= 2.05, J1 = -1.6 cm-1 and J2 = 1.7 cm-1 with R =
2.2x10-3, where the negative sign indicates an antiferromagnetic interaction. We also fitted
The best fit was found for the average
g
the magnetic data using a simpler model assuming J1
=
J2, however it gave a poor quality fit
of the data.
19

Figure 3. Temperature dependence of
χT for 2. The black line represents a fit using the
model described in the text.
ꢀ ꢀ ꢀ ꢀ
ꢀ
2
ꢀ
4
ꢀ
1
ꢀ
2
ꢀ
3
ꢀ
4
H = −J (S • S + S • S ) − J (S • S ) + βgH(S + S + S + S )
(Eq. 2)
1
1
2
3
4
2
Figure 4. Spin topology considered to fit the magnetic data of 2.
20

Two different magnetic couplings ferro- and antiferromagnetic found for
2 are in the
range observed (J = -5.9 cm-1 to J = 2.6 cm-1) for bis(µ-dialkylphosphinato)copper(II)
complexes, coordinated by phosphinate groups in 2.11 (II) mode.[33-37] In fact, the
difference in the magnetic coupling were attributed to ligand-mediated conformational
variations as well as differences on metal environment. Comparing the copper(II) environment of
[Cu(R2PO2)2] where R is the alkyl group, the flattenest tetrahedral environment leads to the
largest O-Cu-O angle as well as the longest intrametallic distance. It is well known that these
features favor the transition from ferromagnetic to antiferromagnetic coupling.[76]
Figure 5
.
Temperature dependence of
χT for 3. The black line represents the best fit using
the parameters described in the text.
The temperature dependence of T for is shown in Figure 5. At 290 K,
K that is smaller than expected for five non-interaction copper(II) ions (2.07
K) with = 2.1. As the temperature decreases, the T decreases down to 0.45
K at 2.0 K. This behaviour indicates dominant intramolecular antiferromagnetic
χ
3
χT is 1.97
cm3
cm3
cm3
·
·
·
mol-1
mol-1
mol-1
·
·
·
g
χ
21

interactions. In order to estimate the magnetic coupling constant the magnetic data was
fitted using the Hamiltonian described in Eq. 3 based on spin topology shown in Figure
6
. All five copper(II) ions are crystallographically independent, so it is expected to have
different magnetic exchange between them. However, Cu1 and Cu2 and Cu3 have similar
metal environment with similar intramolecular distances between them, and also similar
Cu-O-Cu angles. Then, an approximation was made considering the magnetic couplings
between these three ions (J1) equal. Similar approximation was used to the magnetic
coupling (J2) between Cu2···Cu4, Cu3···Cu5 and Cu4···Cu5 although not all of them
have the same metal environment. Due to larger distances between Cu1 and Cu4 as well
as between Cu1···Cu5, Cu2···Cu5 and Cu3···Cu4, the magnetic interactions between
them are expected to be weaker and were neglected to prevent overparametrization.
Firstly, the simplest model was applied considering J1 = J2, but a poor quality fit was
obtained indicating that not all magnetic couplings are the same. Thus, the magnetic data
was fitted considering two distinct coupling constants and the best fit parameters found
were
g
average = 2.25, J1 = -89 cm-1 and J2 = -25 cm-1 with R = 7.8x10-3. There is a good
agreement between experimental data and calculated curve using this model above 20 K.
We stress that a fit performed by considering the magnetic coupling constant between
Cu4···Cu5 is not the same as J2, adding a third magnetic coupling constant (J3). However,
the quality of the fit was not improved by adding one free variable. Although pentanuclear
rings can exhibit a geometric spin frustration,[77] its effect on magnetic susceptibility data is
in general less evident than in cyclic trinuclear ones.
22

ꢀ ꢀ
ꢀ ꢀ ꢀ ꢀ
ꢀ ꢀ
ꢀ ꢀ
ꢀ ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
H = −J1 (S1.S2 + S2 .S3 + S1.S3 ) − J2 (S2 .S4 + S3.S5 + S5 .S4 ) + βgH(S1 + S2 + S3 + S4 + S5 )
(Eq. 3)
Figure 6. Spin topology considered to fit magnetic data of 3.
The temperature dependence of the
χT product for compound 4 is shown in Figure 7.
The
χ
T value at 280 K is 2.15 cm³
·
mol-1
·
K, which is smaller to the calculated one for six
mol-1
non-interacting copper(II) ions with
g
= 2.10 (2.48 cm³
·
·K). Upon cooling, χT value
decreases continuously to 0.29 cm³
·
mol-1
·K due to the presence of predominant
antiferromagnetic interactions. In order to estimate the magnetic coupling constants
between the metal ions, a careful analysis of the distances between paramagnetic centers
was performed based on the mainly pathways for the interactions. As discussed
previously, the crystal structure consists of two Cu3 triangles linked by dppi groups.
Magnetic interactions pathways should be considered between the copper(II) ions from
two the Cu3 units bridged by the dppi groups. In this case, the magnetic interaction via
dppi is expected to be weaker than those ones within the Cu3 triangles, as reported
elsewhere for other phosphinate-based compounds.[78] The first model used to fit the
23

magnetic data consisted in a system described by two magnetic couplings, namely J1
(within the Cu3 triangles) and J2 (between two Cu3 triangular units through dppi bridging
ligands). Nevertheless, no acceptable fit could be obtained for this model below 50 K.
Since there are many possible pathways for magnetic interactions between two Cu3 units
leading to a complicated model described by different magnetic interactions (J), we
considered a simplified model to prevent overparametrization. This model assumes
magnetic interactions within each Cu3 units (J1), and in order to take account the magnetic
interactions between Cu3 units, the average of the magnetic interactions between Cu3 units
was obtained using a mean field approximation (J’
)
(Eq. 4). The best fit parameters were
found for the average g = 2.09, J = − = −
38 cm-1 and J’ 8 cm-1 with R = 2.6x10-2 (solid line
Figure 7).
Figure 7
. Temperature dependence of χT for 4. The black line represents the best fit
using the parameters described in the text.
H = Hexch. +H zeeman
(Eq. 4)
24

ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
exch. = −J(S1 ⋅ S2 + S2 ⋅ S3 + S1 ⋅ S3 + S4 ⋅ S5 + S5 ⋅ S6 + S4 ⋅ S6 )
H
ꢀ
ꢀ
ꢀ
Hzeeman = βgB(S1 + S2 + S3 + S4 + S5 + S6 )
It is known that the magnetic exchange is strongly influenced by the distance between
copper(II) ions, Cu-ligand bond lengths since copper(II) ions can exhibit Jahn-Teller distortion
as well as Cu-O-Cu bond angle. For 2, the copper(II) ions are bridged coordinated by dppi
ligands in 2.11 (II) mode leading to weak ferro- and antiferromagnetic interactions in agreement
with the magnetic coupling found for other complexes having the same bridging mode.[33-37]
The strength of magnetic couplings found in 1, 3 and 4 are larger than the values found for 2. In
fact, the former complexes have shorter Cu···Cu intramolecular distances and additional
bridging ligands such as hfac and oxo when compared with the latter one. The shorter
intramolecular distances between Cu(II) ions and additional bridging ligands favor a better
pathway for magnetic interactions leading to stronger magnetic interactions in 1, 3 and 4 when
compared with 2. Since the magnetic interaction through bridged-phosphinate pathway is weak,
the magnetic interaction through other groups such as µ-oxo or µ-hydroxo bridge is expected to
be dominant.[75,79] Although the magnetostructural correlation in copper(II) bridged-
phosphinate complexes related to Cu-O and Cu-O-Cu is not well understood, such correlation
has been described for other systems containing µ3-oxo bridge.[79] Surprisingly, practically no
dependence of the magnetic exchange interaction on Cu…Cu distance, Cu-O bond length and Cu-
O-Cu bond angle was found for µ3-oxo copper(II) complexes. However, a strong dependence of
the magnetic interaction on the distance between the oxygen atom of µ3-oxo and the plane
containing the three copper(II) ions was observed.[79] Ferromagnetic interactions were found for
25

distances larger than 0.3 Å while antiferromagnetic (AFM) interactions were observed for
distances smaller than 0.3 Å.
In compound 3-4, even if the distance between the plane containing the three copper(II) ions
and µ3-oxo atom is ~ 0.5 Å, the interaction is in contrast with expected one.[79] On the other
hand, AFM interactions were observed for µ-OH bridge,[75] in agreement with the result found
for compound 1. It should also be considered that the presence of µ-hfac and µ2-oxo as bridging
ligands can influence the magnetic interaction making hard the interpretation of the magnetic
properties based on structural parameters.
3.4. Electrochemistry of Copper Complexes
The redox properties of complexes 1 and 2 have been compared with those of the starting
reagent [Cu(hfac)2]. In this experiment, a 10 mL three electrode cell (Ag|AgCl as reference
electrode and a Pt wire counter electrode) filled by a 0.1 M Bu4NPF6 THF solution was used.
Cyclic voltammograms were recorded at 25 mVs-1, being given in the Supporting Information.
The first CV scan on [Cu(hfac)2] (Figure S4) reveals only an irreversible reduction at -0.25 V
that can be assigned to the one step CuII/Cu° reduction since copper deposition on the electrodes
was observed. A reverse scan (Figure S5) shows an irreversible oxidation at +1.0 V that arises
from the copper stripping from the electrode due to the Cu°/CuI oxidation, with possible
formation of [Cu(THF/H2O)4]+ species, as observed by D. Cook et al. for [Cu(hfac)2] in
acetonitrile solution.[80] The CV of 1 shows only two irreversible oxidation processes at +0.85
V and +1.4 V, respectively (Figure S6). The first oxidation can be assigned to the CuII/CuIII
redox couple and the second one to some decomposition of the complex due mainly to the hfac
ligand.[81] In the voltammogram of 1, no CuII/CuI reduction was observed leading to the
26

conclusion that 1 is more suitable to stabilize the CuIII state when compared to [Cu(hfac)2]. The
value of oxidation redox potential of 1 corresponding to the CuII/CuIII process is similar to other
Cu3 complexes (+0.81 V vs. Ag|AgCl normalizing against SCE values).[82] The initial CV scan
of the complex 2 shows two irreversible features at +0.65 V and at -0.8 V, attributed to the
CuII/CuIII and CuII/CuI redox processes, respectively. No deposition of the copper on the
electrodes was observed (Figure S7).
3.5. Catalytic Oxidation of 3,5-DTBC
Scheme 4. Catalytic oxidation of 3,5-DTBC to 3,5-DTBQ in THF.
The catecholase-like activity of polynuclear complexes 1 and 2 has been evaluated using
3,5-DTBC as a model substrate due to its easy oxidation to 3,5-di-tert-butylbenzoquinone (3,5-
DTBQ, Scheme 4). The study was extended to mononuclear [Cu(hfac)2] for comparison, since it
is a simpler source of CuII metal ions. Reactivity studies were performed in air and in dioxygen-
saturated THF solutions (see Supporting Information). By observing the time evolution of the
392 nm absorption band characteristic of the quinone it was ascertained that the three complexes
showed catecholase activity. Figure 8 shows the UV spectral changes of the oxidation catalyzed
by 2. Complexes [Cu(hfac)2] and 1 behave similarly (Figures S15-S16).
Next, the kinetics of the 3,5-DTBC oxidation were determined through the initial rates
method. The pseudo-first-order condition was maintained by using substrate solutions ranging
27

between 10 times and 100 times larger than that of the Cu(II) complex. The initial rates were
determined from the limiting slope of the absorbance versus time curve at t = 0 s using ε(3,5-
DTBQ) = 1720 mol-1·L·cm-1 in THF (Figure S8). The data were analyzed by a Michaelis-
Menten model with a Lineweaver-Burk (double reciprocal) plot (Figures S9-14) to calculate the
kinetic parameters Vmax, Km and kcat (Table 5). The catalytic activity of the complexes under both
experimental conditions (air and saturated dioxygen) decreased in the order [Cu(hfac)2] > 2 > 1.
All complexes showed a higher catalytic activity in dioxygen-saturated solution. The dioxygen
increase the catalytic activity of the complexes usually by reoxidation of the CuI intermediate.
[83,84] [Cu(hfac)2] (kcat 190 h-1) proved to be a better catalysts for 3,5-DTBC oxidation than
some mono- (kcat 4.1 - 63.7 h-1) [85,86] or dinuclear copper complexes (kcat 11.6 - 95 h-1).[87-89]
However, its efficiency is significantly lower than that of highly active Cu2 catalysts (kcat 32400
h-1).[90] As far as we know there are no previous report on the catecholase activity of [Cu(hfac)2]
neither kinetic parameters regarding catecholase activity for copper(II)-based multinuclear
systems coordinated only by oxygen donor atoms. However, when compared with other
trinuclear and tetranuclear copper clusters containing mixed N/O donors coordinated to the metal
ion, complexes 1 (kcat 16.2 h-1) and 2 (kcat 19.4 h-1) act as moderate catecholase-like catalyst.[91,
92] The different catecholase-like activity of the two compounds is certainly related to their
structural features. Several works in the literature illustrate that a number of factors influence the
efficiency of a complex as catalyst in catecholase activity, such as Cu···Cu distance, coordination
geometry of the metal centers, nature of the donor groups and electronic properties of the
ligands, among others.[52-55, 89, 93]
28

Figure 8. Increase of 3,5-DTBQ band at 392 nm (__) in dioxygen-saturated THF for the complex
2 (__) (10-4 mol·L-1) using 1ꢀ10-2 mol·L-1 initial concentration of 3,5-DTBC (__). The spectra were
recorded at 1 min interval.
It is important to note that the molecular structure of complex 1 provides a new model for
multicopper oxidases having three copper active centers that catalyze 4e- reduction of O2,[45, 46]
because it has been found that the enzymatic intermediate must have a fully oxidized Cu3(II) core
where metal ions are mutually bridged by the oxo-independent oxygen atom obtained from
cleavage of the O=O bond.[94]
Table 5. Kinetic parameters for the complexes evaluated
air-saturated solvent THF
Complex
Vmax (Ms-1)
2.9ꢀ10-7
Km (mM)
23.7
kcat (h-1)
10.4
15.8
1
4.4ꢀ10-7
2.59
2
Cu(hfac)2
1.02ꢀ10-6
28.2
36.0
dioxygen-saturated solvent THF
Complex
Vmax (Ms-1)
Km (mM)
58.5
kcat (h-1)
16.2
19.4
4.6ꢀ10-7
5.3ꢀ10-7
5.3ꢀ10-6
1
2
3.89
Cu(hfac)2
126.7
190.2
29