Structural diversity in Cu(i) complexes of the PNP ligand: From pincer to binuclear coordination modes and their effects upon the electrochemical and photophysical properties

Article abstract of DOI:10.1039/c7dt02244a

A set of new copper(i) complexes is synthesized and characterized using a labile PNP pincer ligand (PNP = N,N′-bis(diphenylphosphine)-2,6-diaminopyridine). A homoleptic Cu(i) complex [Cu(PNP-κP¹:κN¹)₂]⁺, (1), was prepared, and taking advantage of the uncoordinated phosphorus atoms in (1), reaction with a second Cu(i) atom bearing secondary ligands (PPh₃, phen or dmp) allows the formation of new complexes: a bimetallic helicate [CuI2(PNP)₂(phen)]²⁺, (2), a mononuclear pincer complex [Cu^I(PNP)(PPh₃)]⁺, (3), and a heteroleptic complex [Cu^I(PNP)(dmp)]⁺, (4). All complexes were characterized by X-ray crystallography, NMR (VT-NMR for (1) and (4)), cyclic-voltammetry, and steady-state and time-resolved luminescence spectroscopy. The fluxional behavior in (1) was studied by ³¹P VT-NMR, where an E_a value of 47.42 kJ mol^-1 was calculated for the intramolecular alternating coordination of -PPh₂ moieties in PNP to the metal atom. This set of compounds reveals the versatility of the PNP ligand when added to the coordinating properties of Cu(i). The four complexes exhibit emission in solution and complexes (2)-(4) display intense luminescence in the solid state. The oscillographic traces showing the decay of the luminescence were fitted to biexponential functions with time constants: 8.0 μs > τ_em,1 > 0.37 μs and 50 μs >τ_{em, 2} > 2.2 μs for complexes (2), (3) and (4), respectively. Radiative relaxation is associated with electronic transitions in both the ligand PNP and metal-to-ligand charge transfer (MLCT).

Full text of DOI:10.1039/c7dt02244a

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Arce, C. M. Vera
, D. L. Escudero, J. Guerrero, A. G. Lappin, A. G. Oliver, D. H. Jara, G. Ferraudi and L. A. Lemus, Dalton
Trans., 2017, DOI: 10.1039/C7DT02244A.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Page 1 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
Structural diversity in Cu(I) complexes of the PNP ligand: from
pincer to binuclear coordination modes and their effects upon
the electrochemical and photophysical properties.
Pablo Arce,† Cristian Vera,† Dayra Escudero,† Juan Guerrero,† Alexander Lappin,§ Allen
§
§
Oliver, Danilo H. Jara,≠ Guillermo Ferraudi, ≠§* Luis Lemus.†*
,
,
†Facultad de Química y Biología, Universidad de Santiago de Chile, Av. Libertador Bernardo
O’Higgins 3363, Estación Central, Santiago, Chile.
§Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556ꢀ
5670, USA.
≠Radiation Research Building, University of Notre Dame, Notre Dame, Indiana 46556, USA.
1

Dalton Transactions
Page 2 of 43
View Article Online
DOI: 10.1039/C7DT02244A
ABSTRACT
A set of new copper (I) complexes is synthesized and characterized using a labile PNP
pincer ligand (PNP=N,N’ꢀbis(diphenylphosphine)ꢀ2,6ꢀdiaminopyridine). A homoleptic Cu(I)
complex [Cu(PNP-
phosphorus atoms in (1), reaction with a second Cu(I) atom bearing secondary ligands (PPh3,
phen or dmp) allows the formation of new complexes: bimetallic helicate
κ κ
P1: N1)2]+, (1), was prepared, and taking advantage of the uncoordinated
A
[CuI2(PNP)2(phen)]2+, (2), a mononuclear pincer complex [CuI(PNP)(PPh3)]+, (3), and a
heteroleptic complex [CuI(PNP)(dmp)]+, (4). All complexes were characterized by Xꢀray
crystallography, NMR (VTꢀNMR for (1) and (4)), cyclicꢀvoltammetry, and steadyꢀstate and
timeꢀresolved luminescence spectroscopy. The fluxional behavior in (1) was studied by 31P VTꢀ
NMR, where an Ea value of 47.42 kJ/mol was calculated for the intramolecular alternating
coordination of ꢀPPh2 moieties in PNP to the metal atom. This set of compounds reveals the
versatility of the PNP ligand when added to the coordinating properties of Cu(I). The four
complexes exhibit emission in solution and complexes (2)ꢀ(4) display intense luminescence in
the solid state. The oscillographic traces showing the decay of the luminescence were fitted to
biexponential functions with time constants: 8.0 ꢁs > τem,1 > 0.37 ꢁs and 50 ꢁs > τem, 2 > 2.2 ꢁs
for complexes (2), (3) and (4), respectively. Radiative relaxation is associated with electronic
transitions in both the ligand PNP and metalꢀtoꢀligand charge transfer (MLCT).
2

Page 3 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
INTRODUCTION
New PNP complexes of Cu(I) have been investigated in this work to establish their solid and
solution phase structures including the energetic parameters associated with the fluxional
behavior and photophysical properties. PNP is regarded as one of various tridentate pincer
ligands, namely ligands whose name is associated with the coordination mode to transition metal
ions. Their general structural features are a central pyridine or phenyl ring orthoꢀdisubstituted by
two coordinating moieties, usually having N or P donor atoms. Since the synthesis of the first
pincer complexes in the early 1970s,1ꢀ3 they have found several applications in chemistry,
including catalysis,4 sensors and the building of supramolecular architectures.5
By far, the chemistry of pincer complexes has been related to the platinum group and metals
such as Ni, Fe and Mo due to their catalytic properties. Complexes of coinage metals have been
received lesser attention. In particular, there is a limited amount of work based on Cu(I)ꢀpincer
complexes.6ꢀ11 In the preparation of the Cu(I) complexes a main difficulty is that the ideal
tetrahedral geometry of the copper(I) ion works against the coplanarity of the donor atoms in
PNP. Therefore, the pincer coordination of PNP or NNN ligands to Cu(I) (P and N are nitrogen
and phosphorus atoms coordinating to copper cation) results in the formation of mononuclear
complexes with the geometry of a highly distorted tetrahedron. In solution phase, the lability of
the d10 copper(I) atom when combined with the structural rigidity imposed by a tridentate pincer
ligand leads to fluxional behavior where the ligand is bidentate with a rapid intramolecular
rearrangement.11, 12 The versatility of pincer ligands with multiple binding sites and PNP, PNN
or NNN as donor atoms, combined with the geometrical preferences of d10 metals10 is a useful
tool in the preparation of new bimetallic complexes with the ligands bridging to both metal
3

Dalton Transactions
Page 4 of 43
View Article Online
DOI: 10.1039/C7DT02244A
centers.13 Although the designation of “pincer” ligand is the result of the mode of coordination to
the metal center, this family of tridentate ligands offers more possibilities of coordination
number, nuclearity and geometries around the metal center.14
With the focus on the preparation of new bimetallic complexes, we have taken advantage of
the uncoordinated phosphorus atoms in [Cu(PNP-κ κ
P1: N1)2]+ (1), Scheme I, to add a second
copper atom that is also coordinated to a secondary ligand; PPh3, phenanthroline or neocuproine.
Using this approach, three new different compounds were obtained under similar synthetic
methods (2) – (4), Scheme I, where the PNP ligand is coordinating in different modes depending
on the other ligand to produce either a bimetallic helicate (2), a mononuclear pincer complex (3)
or a heteroleptic complex (4). The complexes (2), (3) and (4) present an intense luminesce in the
solid phase which make them attractive materials for lightꢀemitting devices. The structures of
this set of complexes are fully characterized by Xꢀray crystallography and in solution by NMR
(VTꢀNMR). The electrochemical behavior is studied by cyclicꢀvoltammetry. In addition,
photophysical properties of the complexes were evaluated both in solution and solid phase, and
are discussed as a function of their structures, opening potential applications in areas such as
luminescent devices, sensors and bioꢀapplications.15,16
4

Page 5 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
Scheme I. Representation of products obtained (2ꢀ4) by direct reaction of the precursor (1) with
heteroleptic Cu(I) complexes of phen, PPh3 and dmp. For complex (4), a second procedure
described in the methodology section can be followed with improved yielding.
5

Dalton Transactions
Page 6 of 43
View Article Online
DOI: 10.1039/C7DT02244A
EXPERIMENTAL SECTION
Materials
Chlorodiphenylphosphine (PPh2Cl), 1,10ꢀphenanthroline (phen), 2,9ꢀdimethylꢀ1,10ꢀ
phenanthroline (dmp) and triphenylphosphine (PPh3) were from SigmaꢀAldrich or Merck and
used as received. Triethylamine (Et3N) was dried by distillation over KOH pellets and other
solvents were dried by their respective procedures. Commercial 2,6ꢀdiaminopyridine was
recrystallized from hot toluene before to use.
The precursor complex [Cu(CH3CN)4](ClO4) was synthesized following the literature
procedure.17
Spectroscopic Measurements.
UVꢀvis and NMR spectra were recorded on solutions deaerated with streams of N2 or Ar gas.
1Dꢀ and 2DꢀNMR spectra were recorded on deuterated solution of the complexes (1,1,2,2ꢀ
tetrachloroethane C2Cl4D2, CD3CN, CDCl3) at 300 K on a Bruker Avance 400 spectrometer
1
operating at 400.13 MHz for H, 100.61 MHz for 13C and 161.97 MHz for 31P nucleus, and
equipped with a 5 mm broadband inverse probe head incorporating a zꢀgradient coil. The
1
chemical shifts (in ppm) for H spectra are reported relative to TMS and calibrated with respect
to the residual solvent signal. 31P{1H} spectra were calibrated with respect to the external
reference H3PO4 10%. The number of scans in each experiment was dependent on the sample
concentration. The unequivocal assignment of the proton signals of the complexes was carried
out by 1HꢀNMR and 2D (COSY, NOESY) experiments.
Electronic spectra were measured on a Scínco Sꢀ3100 spectrophotometer at room temperature
in 1.0 cm quartz cells.
6

Page 7 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
Steadyꢀstate photoluminescence properties were recorded using a Horiba Fluorolog
spectrometer. The solid state emission lifetime measurements were recorded using a Nd:YAG
laser from Spectra Physics (QuantaꢀRay Pro230). The samples were placed in a quartz tube and
irradiated with a 355 nm pump beam (fwhm = 10 ns, 10 Hz repetition rate, 2.5 mJ/pulse). The
photoluminescence of the complexes was measured with a monochromator (Digikrom 240, CVI
Laser Corporation) coupled to a photomultiplier tube and recorded with a 1 GHz Lecroy
oscilloscope. The reaction kinetics were investigated by following the absorbance change at
given wavelengths of the spectrum and incorporating those changes in the % Reaction
dimensionless parameter, ξ = Ιt/Ι0. In ξ, Ι0 is the emission intensity at a zero time and Ιt is the
emission intensity determined at an instant t of the emission decay. The fitting of mono and
biexponential functions to oscillographic traces was done with routines developed in OriginPro
2015. The goodness of the fittings were based in the convergence of the Red. χ2 ≤ 10ꢀ5 and the
Adj. R2~0.99995. Deconvolutions of the emission spectra into Gaussian functions were carried
out using the multipeak fitting routine of OriginPro 2015, which uses similar statistical
parameters for the evaluation of the fitting goodness. In this routine, an excess of contributing
Gausians to the fitting of the spectrum results in an error message while less than the correct
number of Gaussians results in unacceptable values of Red. χ2 and Adj. R2.
Cyclic voltammetry experiments
The Cyclic voltammetry experiments (CV) were performed with a BioꢀAnalytical Systems,
BasiꢀEpsilon electrochemical workstation. The electrolytic cell was a conventional threeꢀ
compartment cell, provided with Pt working electrode, Ag/AgClsat as reference electrode and Pt
as the auxiliary electrode. Measurements were performed at room temperature and N2
atmosphere, using solutions 5x10ꢀ4mol/L of the complexes and 0.10 mol/L of
7

Dalton Transactions
Page 8 of 43
View Article Online
DOI: 10.1039/C7DT02244A
tetrabuthylammonium perchlorate (TBAP) as supporting electrolyte in CH3CN as solvent. The
potentials are presented as E1/2 versus Ag/AgClsat.
Crystallography
Data for compounds (1), (3) and (4) were recorded on a Bruker Kappa APEXꢀII diffractometer
and data for compound (2) was recorded on a Bruker APEXꢀII diffractometer. An arbitrary
hemisphere of data was recorded for each compound at 120(2) K with graphiteꢀmonochromated
MoꢀKα radiation. Data were corrected for absorption and polarization effects and analyzed for
systematic absences to determine the space group. Structures were solved by intrinsic phasing
methods18 and the model developed and refined routinely using leastꢀsquares analysis on all
reflections.19
Except where disorder is present, all nonꢀhydrogen atoms were refined with anisotropic atomic
displacement parameters. Hydrogen atoms were included in idealized positions riding on the
atom to which they are bonded. Uiso(H) was set equal to 1.2 Ueq(C) for methylene and aromatic
H and 1.5 Ueq(C) for methyl hydrogen atoms. Within the structure of compound (1) small
residual electron density was observed near the pendant phosphorus atoms P2 and P4. This was
modeled as partial occupancy oxygen atoms on the supposition that oxidation of these pendant
phosphorus atoms had occurred. Oxygen occupancies were refined to 15% and 18%,
respectively.
Elemental Analysis (EA) was performed to crystalline samples of the complexes. In some
cases, EA results fall within ~2% of error, which can be ascribed to solvent molecules present in
the crystals, as described in the crystallographic section.
8

Page 9 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
N,N’ꢀbis(diphenylphosphine)ꢀ2,6ꢀdiaminopyridine, PNP.
The synthesis of the ligand PNP was carried out according to the reported procedure.14 1.00 g
of 2,6ꢀdiaminopyridine (9.42 mmol) were dissolved in a mixture of 50 mL of anhydrous THF
and 5.25 mL of Et3N (18.85 mmol), and allowed to react for 3 h at room temperature under N2
o
atmosphere in a threeꢀneck reactor. The reaction was cooled to 0 C and 3.46 mL of PPh2Cl
(18.85 mmol) were slowly added, allowing the mixture to reach room temperature and reacting
for 6 h under N2 atmosphere.
The resultant solution was vacuum filtered and rotoꢀevaporated until a brown oil was obtained.
This was dissolved in a minimum amount of CHCl3 and precipitated with diethyl ether. Crystals
of PNP were obtained as colorless needles through diethyl ether diffusion over a concentrated
solution of the ligand. Yield, 65%.
1HꢀNMR (CDCl3): δ(ppm)= 7.44 (m, 8H), 7.35 (t, 13H), 6.48 (d, 2H), 5.00 (d, 2H). 13C NMR
(CD3Cl): δ(ppm)= 157.88, 140.06, 131.73, 129.57, 128.93, 99.47.
[Cu(PNP)2](ClO4), (1)
0.59 g of the precursor complex [Cu(CH3CN)4](ClO4) (1.78 mmol) were dissolved in 40 mL of
a mixture dichloromethaneꢀmethanol 60ꢀ40 in a threeꢀneck reactor, under nitrogen atmosphere.
Then, 1.71 g of the ligand PNP (3.58 mmol) in 40 mL of CH2Cl2 were slowly added dropwise
and the mixture was stirred for 3 h at room temperature under N2 atmosphere. The volume of the
solution was reduced by rotoꢀevaporation and the complex was precipitated with diethyl ether.
Yield 79%. Colorless crystals of (1) were obtained by slow ether diffusion over a CH2Cl2
solution of the complex.
9

Dalton Transactions
Page 10 of 43
View Article Online
DOI: 10.1039/C7DT02244A
Anal. Calcd. for C58H51ClCuN6O4P4 : C, 62.26; H, 4.59; N, 7.51. Found: C, 60.84; H, 4.27; N,
1
7.51. HꢀNMR (CD3CN, 300K): δ(ppm)= 7.43 (t, 1H), 7.34 (t, 4H), 7.22 (t, 8H), 7.01 (s, 8H),
6.31 (d, 2H), 6.00 (s, 2H). 13C NMR (CD3CN): δ (ppm)= (ꢀPy) 142.0, 135.9, 131.0, (ꢀPPh2)
131.4 (d, 2JCꢀP: 18 Hz), 131.2 (overlapped), 130.5 (s), 129.3 (broad).
[Cu2(PNP)2(phen)](ClO4)2, (2)
This complex was prepared from (1) and [Cu(CH3CN)4](ClO4) as precursors. To 0.56 g of the
complex [Cu(CH3CN)4](ClO4) (1.71 mmol) dissolved in a mixture of dichloromethaneꢀmethanol
60ꢀ40 in a threeꢀneck reactor, were added 0.31 g of phen (1.71 mmol) dissolved in 40 mL of
CH2Cl2. Then, 1.91 g of the complex (1) (1.71 mmol) dissolved in the CH2Cl2ꢀmethanol mixture
were slowly added to the reactor and the reaction was kept at room temperature with a N2 stream
for 6 h.
An orange oil was obtained after evaporation of the solvents at reduced pressure. The oil was
precipitated with diethyl ether to produce a yellow powder. Yellow crystals were obtained by
slow ether diffusion over a concentrated solution of the complex in CH2Cl2ꢀmethanol as solvent.
Yield, 83%.
Anal. Calcd. for C70H59Cl2Cu2N8O8P4 : C, 57.50; H, 4.07; N, 7.66. Found: C, 57.20; H, 3.99; N,
7.67. 1HꢀNMR (C2Cl4D2, 300K): δ(ppm)= 9.00 (d, 1H), 8.77 (d, 1H), 8.27 (s, 1H), 8.11 (dd, 1H),
7.96 (s, 2H), 7.66 (s, 3H), 7.38 – 7.25 (m, 5H), 7.25 (s, 1H), 7.16 (m, 6H), 7.07 (t, 2H), 6.97 (t,
13
3H), 6.74 (d, 2H), 6.63 (d, 2H), 6.29 (d, 1H), 5.94 (d, 1H), 4.91 (s, 1H). C NMR (C2D2Cl4):
δ(ppm)= (ꢀPy, phen) 157.9, 154.1, 150.7, 150.1, 143.8, 142.4, 139.1, 106.4, 101.5, (ꢀPPh2) 134.1,
133.8, 133.5, 133.3, 133.2, 133.1, 132.5, 131.7, 131.4, 131.1, 130.8, 130.4, 129.9, 129.6, 129.3,
128.5.
10

Page 11 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
[Cu(PNP)(PPh3)](ClO4), (3)
0.53 g of the ligand PNP (1.11 mmol) dissolved in CH2Cl2 were slowly added to 0.362 g of the
complex [Cu(CH3CN)4](ClO4) (1.11 mmol) dissolved in 40 mL of a mixture CH2Cl2ꢀmethanol
60ꢀ40, contained in a threeꢀneck reactor under N2 atmosphere. After 30 min, 0.29 g of PPh3 (1.11
mmol) in CH2Cl2 were added to the reactor and the reaction was keep under N2 atmosphere for 3
h. The volume of the solution was reduced by rotoꢀevaporation and diethyl ether was used to
precipitate the complex as a white powder. Yield, 76%. Colorless crystals were obtained by slow
ether diffusion over a CH2Cl2 solution of the complex.
Anal. Calcd. for: C47H40ClCuN3O4P3 : C, 62.53; H, 4.47; N, 4.65. Found: C, 63.64; H, 4.53; N,
4.68. 1HꢀNMR (C2Cl4D2, 300K): δ(ppm)= 7.57 (t, 1H), 7.33 (t, 8H), 7.18 (m, 21H), 7.06 (t, 6H),
6.50 (d, 2H), 5.95 (d, 2H). 13C NMR (C2D2Cl4): δ(ppm)= (ꢀPy) 155.1, 143.1, 133.9, (ꢀPPh2,
PPh3) 133.6, 133.5, 131.5, 131.4, 130.7, 130.4, 129.1, 102.2.
[Cu(PNP)(dmp)](ClO4), (4)
Two methods were used to prepare complex (4): The first (Scheme I), was by direct addition
of complex (1) to a CH2Cl2 solution containing [CuI(CH3CN)4](ClO4) and dmp in a ratio 1:1.
After purification on a chromatographic column, a considerable amount of oxidized product was
retained by the alumina. The yield of (4) was lower than 40%.
As a second method, in a threeꢀneck reactor were dissolved 0.1349 g (0.41mmol) of the
precursor [CuI(CH3CN)4](ClO4) with 30 mL of CH2Cl2 under N2 atmosphere. Then, 0.4327g of
PPh3 (1.65 mmol) were slowly added to the reactor and the mixture was kept under N2 for 2 h. A
white solid was obtained after rotoꢀevaporation, yielding the complex [CuI(PPh3)4](ClO4). The
11

Dalton Transactions
Page 12 of 43
View Article Online
DOI: 10.1039/C7DT02244A
tetraphosphine complex was dissolved in 30 mL of CH2Cl2 and a solution containing 0.1969 g
(0.41mmol) of the ligand PNP was added. After 2 h, the solution was rotoꢀevaporated and the
resulting white solid was washed with diethyl ether in a Soxhlet system to remove uncoordinated
phosphine. After dissolution of the solid in CH2Cl2, 0.085 g of a neocuproine (0.41 mmol)
solution were slowly added to the reactor and the yellow solution was stirred for 2 h under N2.
An orange oil was isolated after rotoꢀevaporation. Several washes with diethyl ether to remove
the free phosphine which was displaced by the ligand dmp, resulted in the precipitation of the
complex [Cu(PNP)(dmp)](ClO4). Yield: 73%.
Anal. Calcd. for C43H37ClCuN5O4P2: C, 60.85; H, 4.39; N, 8.25. Found: C, 61.23; H, 4.35; N,
1
8.35. HꢀNMR (C2Cl4D2, 300K): δ(ppm)= 8.37 (d, 2H), 7.90 (s, 2H), 7.65 (2, 2H), 7.54 (t, 1H),
7.22 (t, 4H), 7.09 (t, 8H), 6.97 (d, 8H), 6.82 (d, 2H), 6.08 (s, 2H), 2.39 (s, 6H). 13C NMR
(C2D2Cl4): δ(ppm)= (ꢀPy, dmp) 159.8, 142.1, 128.0, 138.6, 136.5, 136.4, 129.7, 126.6, 126.1, (ꢀ
2
3
PPh2) 131.2 (d, JCꢀP: 19.5 Hz), 131.0 (s), 129.3 (overlapped), 129.2 (d, JCꢀP: 8.4Hz), (ꢀCH3)
26.8.
12

Page 13 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
RESULTS AND DISCUSSION
The geometry imposed by the Cu(I) atom allows the coordination of two molecules of PNP
acting as bicoordinating ligands instead of the pincerꢀlike triꢀcoordination. Consequently, there
are two donor atoms that are able to bind a second copper atom. The complex thus formed,
[Cu(PNP)2]+ (1), can be used as a precursor for new bimetallic complexes. On the basis of this
proposal, the bimetallic complex [CuI2(PNP)2(phen)]2+ (2) was prepared by mixing equimolar
amounts of [Cu(PNP)2]+ and [Cu(phen)(CH3CN)2]+, Scheme I.
Under similar experimental conditions used for the preparation of (2), the syntheses of other
two bimetallic compounds were attempted. One derived from triphenylphosphine
([CuI2(PNP)2(PPh3)2]2+) and another derived from neocuproine ([CuI2(PNP)2(dmp)]2+). However,
the addition of equimolar amounts of [Cu(PPh3)2(CH3CN)2]+ to the solution containing complex
(1) leads to the rearrangement of both complexes and the formation of two equivalents of a new
monometallic complex, [CuI(PNP)(PPh3)]+ (3), Scheme I. This is most likely due to the steric
hindrance produced by the triphenylphosphine ligand, which is bulky enough to prevent the
coordination of two PPh3 to the same cooper atom. Complex (3) has an unusual geometry with
PNP tricoordinated as a pincer ligand. This complex can also be synthesized in higher yields
mixing Cu(I), PNP and PPh3, in a ratio 1:1:1, as described in the experimental section.
The preparation of the neocuproine binuclear complex was attempted in a reaction where
[Cu(PNP)2]+ and [Cu(dmp)(CH3CN)2]+ were mixed in stoichiometric amounts. The reaction
yield the heteroleptic complex [CuI(PNP)(dmp)]+ (4), and oxidation products which were not
studied in this work. Despite the fact that phen and dmp have similar structures, models suggest
that the steric hindrance caused by the 2,9ꢀdimethyl substituents in dmp causes steric congestion
with the diphenylphosphine groups that impedes the formation of a bimetallic species. The yields
13

Dalton Transactions
Page 14 of 43
View Article Online
DOI: 10.1039/C7DT02244A
in the synthesis of complex (4) can also be improved by the separate method described in the
experimental section.
The set of complexes of PNP and secondary ligands such as PPh3 or dmp obtained in this work
reveal the versatility of the PNP ligand when it is coordinated to Cu(I). Indeed, it is able to
function in a pincerꢀfashion, as a bicoordinating ligand and as a tricoordinating ligand to form
bimetallic species.
Structural characterization in solution and solid phase.
Crystallographic structures.
The crystal structures (1) ꢀ (4) are analyzed in terms of the geometrical parameter, τ4, for fourꢀ
coordinated complexes proposed by Yang et al.,20 which uses the bond angles of the primary
coordination sphere of the metal atom to establish a numerical value related to geometry. The
values of τ4 will range from 1.00 for a perfect tetrahedral geometry, to zero for a perfect square
planar geometry. Intermediate structures, including trigonal pyramidal and “seesaw”, fall within
the range 0.00 to 1.00.
The complex [Cu(PNP)2](ClO4) (1) crystallizes as yellow blockꢀlike crystals. There are four
molecules of the copper cation and associated perchlorate anion in the unit cell of the monoclinic
space group P21/n. The complex consists of a copper coordinated by one phosphorous and the
pyridine nitrogen of each PNP ligand, leaving one pendant uncoordinated phosphorus on each
PNP ligand, Fig. 1. The copper atom has a τ4 value of 0.72, indicative of a seesaw geometry. A
small, yet significant amount, of residual electron density located about 1.4 Å from both pendant
phosphorus atoms was observed in a difference Fourier map. Given the propensity for
14

Page 15 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
phosphines to form phosphine oxides, it is not unreasonable that a similar process occurred here
and that there is a small percentage of phosphine oxide present in the lattice.
Figure 1. Molecular structure of [Cu(PNP)2](ClO4), (1).
The bond distances and angles between Cu and the coordinating atoms in both PNP ligands
are similar: The distance from Cu(1) to N(2) and N(5) are almost identical, 2.069 and 2.059 Å,
respectively. The same for Cu(1)ꢀP(1)/P(3), with distances of 2.209 and 2.222 Å. Also, angles
formed in the coordinative environment of the copper atom are similar, P(1)ꢀCu(1)ꢀ(N2) =
85.95O and P(3)ꢀCu(1)ꢀN(5) = 86.07O.
The synthesis of the complex [Cu(PNP)2]+ with iodide as counterion (the same PNP ligand)
has been already reported by Pan et al.12 from the reaction of copper(I) iodide with PNP in a
molar ratio 1:2. While crystals were obtained, the quality of the crystals prevented these authors
from making a full refinement of the structure. In our treatment of [Cu(PNP)2]ClO4 crystals of a
sufficient quality were obtained for a full refinement of the structure.
15

Dalton Transactions
Page 16 of 43
View Article Online
DOI: 10.1039/C7DT02244A
Compound
Empirical formula
Formula weight (g/mol)
Temperature (K)
Cryst. Syst., space group.
Unit cell dim. (Å, deg)
a (Å)
(1)
(2)
(3)
C47H40ClCuN3O4P3
902.72
(4)
C43H36ClCuN5O4P2
847.70
C58H50ClCuN6O4.34P4
1123.35
C71H61Cl3Cu2N8O8.50P4
1519.58
120(2)
120 (2)
120(2)
120(2)
Monoclinic, P21/n
Triclinic, P-1
Monoclinic, P21/n
Triclinic, P-1
19.6651(16)
12.1590(11)
22.774(2)
90°
12.8814(18)
14.442(2)
10.4216(9)
30.083(3)
13.3873(11)
90°
9.5788(11)
13.3745(16)
17.068(2)
105.687(3)°
93.029(3)°
101.325(3)°
2051.1(4)
2
b (Å)
c (Å)
22.538(3)
107.055(2)°
94.260(2)°
111.323(2)°
3656.2(9)
α
, deg.
, deg.
, deg.
104.059(2)°
90°
90.832(2)°
90°
β
γ
volume (Å3)
5282.4(8)
4
4196.6(6)
4
Z
2
Calculated density (g/cm3)
Absorption coeff. (μ) (mm−1
F(000)
1.413
1.380
1.429
1.373
)
0.640
0.838
0.747
0.724
2323
1560
1864
874
crystal size (mm3)
0.22 × 0.15 × 0.08
1.229° to 26.394°
-24 ≤ h ≤ 24, -15 ≤ k ≤
15, -28 ≤ l ≤ 28
63333
0.29 × 0.26 × 0.14
1.568º to 28.305º
-17 ≤ h ≤ 16, -19 ≤ k ≤
19, -26 ≤ l ≤ 30
49882
0.28 × 0.26 × 0.08
1.665° to 26.482°
0.26 × 0.18 × 0.07
1.621° to 26.516°
θ
range (deg)
-13 ≤ h ≤ 13, -37 ≤ k ≤
37, -16 ≤ l ≤ 16
45776
-12 ≤ h ≤ 10, -16 ≤
k ≤ 16, -21 ≤ l ≤ 21
31677
Index ranges
Reflns. collected/unique
Independent reflections
10809 [Rint =
0.0558]
18003 [Rint =
0.0255]
8460 [Rint =
0.0669]
8620 [Rint = 0.0640]
100%
100 %
100 %
100%
Completeness to
θ (%)
Absorption correction
Max and min transmn.
Numerical
1.0000 and 0.9064
10809 / 0 / 677
1.022
Numerical
0.9830 and 0.8902
18003 / 14 / 907
1.027
Numerical
1.0000 and 0.9082
8620 / 0 / 532
1.013
Numerical
0.9795 and 0.9433
8460 / 0 / 505
1.044
Data/restraints/parameters
GOF on F2
R1 = 0.0395,
R1 = 0.0414,
R1 = 0.0369,
R1 = 0.0522,
Final R indices [I >2σ(I)]
R indices (all data)
wR2 = 0.0906
R1 = 0.0617,
wR2= 0.1142
R1 = 0.0538,
wR2 = 0.0753
R1 = 0.0624,
wR2 = 0.1286
R1 = 0.0887,
wR2 = 0.1010
wR2 = 0.1220
wR2 = 0.0849
wR2 = 0.1471
Largest diff peak and hole
(e/Å3)
0.602 and -0.391
2.318 and -0.770
0.448 and -0.514
0.647 and -0.951
Table 1 : Crystallographic data and refinement.
16

Page 17 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
Aside from the parameters communicated in Table I, our observations and those of Pan et al.
are in agreement. According to these results, [Cu(PNP)2]I crystallizes in the same space group
(monoclinic P21/n), where the copper cation is tetracoordinated in a tetrahedral geometry by the
pyridine nitrogen and one –PPh2 of each PNP ligand, leaving two pendant diphenylphosphine
groups in the molecule. In addition, the number of molecules (Z=4) in the unit cell and the cell
parameters of both structures are very similar (Table 1), which may indicate that the solid
structures of both the cation are closely related in the perchlorate and iodide complexes.
Figure 2. Molecular structure of [Cu2(PNP)2(phen)](ClO4)2, (2).
17

Dalton Transactions
Page 18 of 43
View Article Online
DOI: 10.1039/C7DT02244A
The complex (2) crystallizes as pale yellow blockꢀlike crystals from a dichloromethane/ethanol
solution. There is one molecule of the bimetallic complex with two Cu(I) cations, two associated
perchlorate anions, one half dichloromethane and five molecules of ethanol within the
asymmetric unit of the triclinic space group Pꢀ1. This copper complex, Fig. 2, consists of two
copper centers with distorted tetrahedral coordination geometry, where Cu(1) is coordinated by
the two N atoms of a phenanthroline ligand (Cu(1)ꢀ N1/N2=2.078(2)/2.091(2) Å) and the
terminal phosphorus atoms of two PNP ligands (Cu(1)ꢀP1/P3=2.2686(7)/2.2481(6) Å). The τ4
value is 0.82 for this copper, indicating a slightly distorted tetrahedral environment. The second
copper is coordinated by the two pyridine nitrogen atoms (Cu(2)ꢀN4/N7=2.077(2)/2.078(2) Å)
and the two remaining phosphorus atoms of the PNP ligands (Cu(2)ꢀP2/P4=2.2135(7)/2.2142(8)
Å). The geometry of Cu(2) deviates more from an ideal tetrahedron and is close to a seesaw
geometry, τ4 = 0.73. In addition, the wrapping of the PNP ligands around the copper atoms
determines the helical nature of the complex, where both M and P enantiomers crystallize as a
racemic mixture in the unit cell, Fig. 3.
Figure 3. Molecular structure of enantiomers Mꢀ(2) (left) and Pꢀ(2) (right). PNP ligands are
colored to highlight the helicity of the enantiomers.
18

Page 19 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
The preference of the copper (I) atom in adopting a tetrahedral geometry, and the flexibility of
PNP when acting as bicoordinating ligand, are factors that favor helicity in the bimetallic
complex. A third feature, a ππꢀinteraction between the pyridine ring of one PNP and the phenyl
ring of the another PNP also contributes to the formation of this bimetallic helicate. These ππꢀ
interactions are characterized by centroid to centroid distances of 3.882 and 3.695 Å. Previously,
we have observed 21 that the folding of the ligands upon coordination to yield a helicate complex
allows the establishment of ππꢀinteractions capable of stabilizing the helicate conformation.
These are not feasible when the conformation is that of a mesocate complex. This complex is
added to the few examples of Cu(I) bimetallic helicates derived of pincer ligands. 22, 23
Complex (3) crystallizes as yellow, tabletꢀlike crystals. There are four molecules of the
complex and associated perchlorate anions in the unit cell of the centrosymmetric, monoclinic
space group P21/n. The copper center is coordinated in a fourꢀcoordinate “butterfly” or seesaw
fashion by the nitrogen and phosphorus atoms of the PNP pincer ligand and the phosphorus of a
triphenylphosphine ligand, Fig. 4.
Figure 4. Molecular structure of [Cu(PNP)(PPh3)](ClO4), (3).
19

Dalton Transactions
Page 20 of 43
View Article Online
DOI: 10.1039/C7DT02244A
Inspection of the bond angles about the Cu center shows the highly distorted nature of the
coordination geometry. The τ4 calculation yields a value of 0.67, which is intermediate between
square planar (τ4 = 0) and tetrahedral (τ4 = 1). Presumably, the steric bulk of the
triphenylphosphine ligand enforces such distortions. There is some degree of twisting involved
in the PNP ligand coordination. The sum of the angles involving N2ꢀCu1ꢀP1/2 is 162.78° while
the P1ꢀCuꢀP2 bond angle is only 139.85(2)°. The complex forms oneꢀdimensional Hꢀbonded
chains in the solid state. The chains are propagated by hydrogen bonds from the amine nitrogen,
N1 to a perchlorate oxygen (O1) and from the amine, N3, to a perchlorate oxygen (O2’) related
by the crystallographic cꢀglide plane.
Recently, Richeson et al.10 reported two neutral copperꢀpincer complexes, [2,6ꢀ
{Ph2PNMe}2(NC5H3)]CuBr (A, Scheme II) and [2,6ꢀ{Ph2PNMe}2(NC5H3)]CuOTf (B, Scheme
II) ([2,6ꢀ{Ph2PNMe}2(NC5H3)] = N,N´ꢀbis(diphenylphosphine)ꢀ2,6ꢀdi(methylamino)pyridine
and OTF ( = OSO2CF3), where the pincer ligand is triꢀcoordinated at the copper center in a
similar manner to (3), with bromine or OTF as the fourth ligand in a geometry described as
trigonal pyramidal (τ4 = 0.78 for A and τ4 = 0.75 for B). Most of the bond distances between the
pincer fragment and the copper atom in A and B are smaller than in (3). For complex A, CuꢀP
distances are 2.2245 and 2.2386 Å and the CuꢀNpy distance is 2.1230 Å, while in complex B the
CuꢀP distances are 2.2264 and 2.2301 Å and the CuꢀNpy distance is 2.0574 Å, which is the
slightly shorter than that in cationic complex (CuꢀNpy = 2.070(2) Å) (3).
20

Page 21 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
Scheme II
Interestingly, the CuꢀPNP fragment in (3) matches closer to another complex reported by
Hayashi et al.,9 [Cu(PNP)(CH3CN)]PF6, (C, Scheme II) where the PNP is a 2,6ꢀBisꢀ
(phosphaethenyl)pyridine ligand with a 2,4,6ꢀtriꢀtꢀbutylphenyl substituent (Mes*) at each
phosphorus atom. Only small differences in the bond distances for the CuꢀPNP fragments of both
complexes are found: For complex C, CuꢀP distances are 2.313(2) and 2.303(1) Å and CuꢀNpy
distance is 2.071(4) Å. The PꢀCuꢀP angle is slightly larger in C (141.43o), ascribed to the smaller
steric hindrance of the MeCN ligand compared to the bulky PPh3 which forces to the Cu atom to
move away from the PꢀNpyꢀP plane in (3). In addition, the phosphaethenyl complex has the same
τ4 value that (3) (τ4 = 0.67), both complexes having a seesaw geometry. 24
Despite the similarity between the PNP ligands in A, B and (3) (a methyl substituent instead H
at the nonꢀcoordinating amines), bond distances, angles and geometry around the copper atom
are almost identical to the complex C, where the PNP ligand is substantially different. On this
basis, it is tempting to relate this to the cationic nature of complexes [Cu(PNP)(PPh3)]+ and C, in
contrast to the neutral complexes A and B.
21

Dalton Transactions
Page 22 of 43
View Article Online
DOI: 10.1039/C7DT02244A
Figure 5. Molecular structure of [Cu(PNP)(dmp)](ClO4), (4).
The complex (4) crystallizes as yellow blockꢀlike crystals from an ethanol solution. There are
two molecules of the Cu cation and associated perchlorate anion in the unit cell of the
centrosymmetric triclinic space group Pꢀ1. The Cu is coordinated in a distorted tetrahedral
fashion by the two nitrogen atoms of the dmp ligand and the pyridine nitrogen and one
phosphorus of the PNP ligand, Fig. 5. The remaining phosphorus of the PNP ligand is not
coordinated. The τ4 value for the Cu center is 0.70, indicative of a seesaw geometry. The Cuꢀ
Ndmp distances are all similar (2.05 – 2.08 Å) and the CuꢀP bond distance is 2.1876(10) Å. The
amide N3, forms a hydrogen bond to one perchlorate oxygen, while the amide N5 is oriented
towards the fiveꢀmembered chelate ring formed by the phenanthroline and the copper.
22

Page 23 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
NMR characterization and fluxional behavior
A number of NMR experiments were performed to assess both structural and redox stability of
the copper complexes in solution. In general, well defined proton signals are observed for all the
compounds, indicating that no oxidation to Cu(II) occurs under the experimental conditions. VTꢀ
NMR experiments were carried out for complexes (1) and (4), where there is the possibility of
fluxionality based on their crystal structures and other information.
The 1ꢀD (1H, 31P) and 2ꢀD NMR spectra of the homoleptic complex (1) were recorded in
31
deuterated acetonitrile as solvent (Figs. S1, S2). In P NMR, the observation of only one signal
of coordinated PNP contrasts with expectations from the crystal structure of this complex, where
only one diphenylphosphine moiety is coordinated to the copper, for each PNP ligand. No
significant 31P signal corresponding to P(V) from phosphine oxide was detected around 20.3
ppm.
The halfꢀwidth of the unique broad signal centered around 28.0 ppm to phosphorus
(ꢀν1/2=213.5 Hz at 300K) is considerably broader than the free PNP ligand (ꢀν1/2=5.7 Hz),
suggesting a fluxional behavior in solution, where both phosphine moieties of each PNP ligand
are alternating for the copper coordination, at rates comparable to the detection rate of the
instrument.
In order to determine energetic parameters involved in such an intramolecular exchange, 31P
VTꢀNMR experiments for (1) were performed in CD3CN in the range 228ꢀ256 K, Fig. 6. At
temperatures below 240 K, the 31P signal splits in two broad singlets, corresponding to two
coordinated and two uncoordinated phosphorus atoms, while complete coalescence occurs at Tc≈
244 K. Rate constants for the phosphorus exchange were obtained through the lineꢀshape
analysis by fitting to Bloch derived equations, Fig. 6.
23

Dalton Transactions
Page 24 of 43
View Article Online
DOI: 10.1039/C7DT02244A
256'K
!
!
248'K
244'K
240'K
!
!
!
!
!
236'K
232'K
!
!
!
!
228'K
5000
4500
4000
3500
ν
(Hz)
31
Figure 6. P VTꢀNMR of (1) in CD3CN at temperature range 228ꢀ256 K. The simulated bands
by lineshape analysis are represented by red curves. Rates constants (sꢀ1) : 1.76x103, 2.28x103,
3.33 x103, 4.78 x103, 5.78 x103, 1.52 x104, 2.41 x104 at 228 K, 232 K, 236 K, 240 K, 244 K, 248
K and 256 K, respectively.
24

Page 25 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
An activation energy value of Ea=47.42 kJ/mol was determined from an Arrhenius plot (ln k vs
1/T) and the activation parameters were derived from an Eyring plot are: ꢀH≠=45.42 kJ/mol,
ꢀS≠=17.40 J/K mol, ꢀG≠=40.23 kJ/mol (calculated at 298K), Fig. S3. Activation energy values
can be compared with those of complexes undergoing a similar intramolecular exchange. For
instance, Abel et al.25 reported the fluxional behavior of terpyridine when is acting as a
bicoordinating ligand in the square planar complexes [M(C6F5)2(terpy)], namely Ea=75.7,
ꢀH≠=72.8 kJ/mol, ꢀS≠=5.8 J/K mol, ꢀG≠=71.0 kJ/mol for M=Pd and Ea=77.0, ꢀH≠=74.1 kJ/mol,
ꢀS≠=10.8 J/K mol, ꢀG≠=70.9 kJ/mol for M=Pt.
In the Pd and Pt complexes, the alternating coordination of the nonꢀcentral pyridine rings in
terpy must occur in a squareꢀplanar geometry, where steric effects can further influence the
dynamics. A lower activation energy value for the fluxional process of (1) in tetrahedral
geometry can be expected as a result of the lability of Cu(I) given the lack of crystal field
stabilization energy in the d10 configuration.26
A smaller activation energy value (Ea=33 kJ/mol in THFꢀd8) accounting for a more rapid
intramolecular arrangement was reported by Petrovic et al.11 for the similar Cu(I)ꢀpincer
complex,
[(TLtBu)CuBr],
where
TLtBu=2,6ꢀbis[(1,3ꢀdiꢀtertꢀbutylimidazolinꢀ2ꢀ
imino)methyl]pyridine. This complex crystallizes with the ligand coordinated in a chelating
fashion [(TLtBu- N1: N2)CuBr], which leaves the copper(I) atom triꢀcoordinated in a trigonalꢀ
κ
κ
planar geometry, whereas in solution the coordinated and uncoordinated arms of the ligand can
be differentiated by NMR at temperatures below Tc=176 K.
25

Dalton Transactions
Page 26 of 43
View Article Online
DOI: 10.1039/C7DT02244A
According to its crystal structure, complex (2) corresponds to a bimetallic complex with both
copper atoms in different chemical environments, resulting in the loss of symmetry of the PNP
ligand and a more complex pattern of proton signals in the NMR spectra. For complex (2), the
numbering of the protons of the PNP and phen ligands is shown in Fig. S4 (NMR spectra in Figs.
S5ꢀS8 and assignments in Table S1). For the phenanthroline coordinated to Cu(1) (Fig. 2), the 1H
signals are those expected for a symmetrically coordinated phenanthroline ligand suggesting that
the compound in solution has a symmetry axis crossing both copper atoms and bisecting the
phenanthroline ligand.
The shielding effect of the phenanthroline ligand shifts to higher field the protons of the ꢀPPh2
groups coordinated to Cu(1), where they appear as a double set of wellꢀdefined signals,
stemming from the different chemical environments perceived by the protons pointing to the
phen ligand on each phenyl ring. By contrast, the protons of the –PPh2 groups coordinate to
Cu(2) appear at lower field and the signals have lower resolution, due to lower steric hindrance
and a greater possibility that molecular motions of the phenyl rings, result in broader signals.
From the 1HꢀNMR and NOESY spectra, it is possible to conclude that this complex retains its
helical nature in solution and its conformation must be very similar to that observed in the crystal
structure. For example, the diastereotopic behavior of protons on both phenyl rings in the couple
of phosphines coordinated to Cu(1) (6.5 ꢀ7.0 ppm, Fig. S5) is indicative of helicity of the
complex in solution phase. Also, a NOESY interaction between H3’ on the pyridine rings (5.94
ppm) and H2, H9 on the phenanthroline ligand (9.0 ppm) is only possible when the complex has
a folding way as that observed in its crystal structure (Fig. S8). In fact, according to the crystal
structure, the distance between both interacting protons is ~2.5 Å, which is in the range of
distances that NOESY interactions are detected.
26

Page 27 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
In the 31P spectrum of the complex (2) in C2Cl4D2 (Fig. S7), two singlet signals at δ=42.14
ppm (ꢀν1/2=73.76 Hz) and δ=31.20 ppm (ꢀν1/2=57.53 Hz), are observed in agreement with the
symmetry of the molecule in solution where both –PPh2 groups coordinated to Cu(1) are
equivalent to each other as are the two phosphine groups coordinated to Cu(2). No significant
variation associated to fluxionality of the complex was observed by recording the 31P spectrum at
different temperatures.
The monometallic and heteroleptic nature in solution of the complex (3) was unequivocally
confirmed by the concerted use of 1ꢀD and 2ꢀD NMR techniques using C2Cl4D2 as solvent (Figs.
S9ꢀS10). The integration of the proton signals agrees with one PNP and one PPh3 ligand, both
shifted respect to the free ligands. The 31P spectrum consist of two phosphorus signals coupled to
each other: A doublet at δ=40.40 ppm, corresponding to two P atoms of PNP ligand, and a triplet
at δ=3.58 ppm corresponding to one P atom of the PPh3 ligand. Narrow signals in the 31P
spectrum recorded at room temperature agree with no fluxional behavior in solution.
1
The HꢀNMR spectrum of (4) in C2Cl4D2 (Fig. S11) shows the pattern of signals of the dmp
and PNP ligands, both shifted to low field relative to the free ligands. In solution of deuterated
acetonitrile, an intramolecular exchange process similar to the described for the complex (1) is
expected because, as observed in the crystal structure of (4) (Fig. 6), one diphenylphosphine
group in PNP is not coordinated to the copper atom.
The 31P spectrum of (4) at 300 K consists of an unique broad signal at 23.2 ppm (ꢀν1/2=135.9
Hz) in accord with an intramolecular exchange process. However, in VTꢀNMR experiments
31
performed between 280 and 224K (Fig. S12), the splitting of the P signal was not observed.
27

Dalton Transactions
Page 28 of 43
View Article Online
DOI: 10.1039/C7DT02244A
Furthermore, the peak width varies only slightly in the range of temperatures measured,
ꢀν1/2=136 Hz and 177 Hz at 224 and 280K, respectively. At the lower temperatures (224 and
31
228K), a shoulder is observed at high field with respect to the main P peak, suggesting that at
temperatures below 224K both phosphorus signals might be resolved and the intramolecular
exchange would occur at rates lower than the detection rate of the instrument.27
From the simplest analysis of the fluxional behavior in (4) and (1), it can be concluded that the
phosphine moieties in (4) are exchanging at rates higher than those in the homoleptic complex
(1). However, in complex (1) there are four phosphorus atoms exchanging for the copper
coordination, whereas in complex (4) only two phosphines are alternating the coordination.
Taking this into account, it is not surprising that the activation energy for the fluxional process in
(4) should be smaller than the value of (1).
Redox behavior of Cu(II)/Cu(I) couples.
The redox behavior of the set of complexes was investigated by cyclic voltammetry, given the
dependence of the coordination geometry around the copper atom, on the potentials of the
CuII/CuI couples. A general experimental observation is that oxidation of Cu(I) is difficult when
complexes are bearing bulky ligands, considering the change of geometry upon oxidation
(flattening of coordinative environment), shifting the oxidation potentials toward more positive
values. Conversely, Cu(I) centers can be more facile oxidized when complexes have extended
aromatic ligands, for example phen or dmp ligands in comparison to the phosphine moieties of
PNP. Cyclic voltammetry experiments were carried out with CH3CN solutions containing 5x10ꢀ4
M of the complexes and 0.10 M of the TBAP supporting electrolyte. An Ag/AgClsat electrode
was used as the reference electrode.
28

Page 29 of 43
Dalton Transactions
View Article Online
DOI: 10.1039/C7DT02244A
Table 2. Cyclic voltammetry results for (1)ꢀ(4) in CH3CN and Ag/AgClsat as reference.
Complex
(1)
Eap(mV)
758
Ecp(mV)
615
E1/2 (mV)
687
ꢁEp(mV)
143
Cu(1) 345
Cu(2) 680
682
Cu(1) 276
Cu(2) 548
576
310
69
(2)
614
138
629
106
(3)
(4)
390
301
345
89
Despite the relationship between coordination geometry and oxidation state for copper
complexes, a clear relationship between the coordination environment of the copper atom as
deduced from the comparison of the τ4 values and the potential, Eap, at which oxidation occurs,
was not found, Table 2.
0.8
0.4
0.6
0.0
0.4
-0.4
0.2
0.8
0.0
0.4
-0.2
0.0
-0.4
-0.4
-0.8
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
E/V vs. Ag/AgCl
E/V vs. Ag/AgCl
sat
sat
Figure 7. Cyclic voltammetry of complexes: a) (1), b) (4) and c) (2), at 5x10ꢀ4 M of complex in
CH3CN and scan rates of 50 mV/s, Ag/AgClsat as reference.
29