Synthesis and coordination behavior of CuI bis(phosphaethenyl) pyridine complexes

Article abstract of DOI:10.1039/c1cc11315a

Cu^I complexes bearing BPEP as a PNP-pincer type phosphaalkene ligand undergo effective bonding interactions with SbF₆^- and PF₆^- as non-coordinating anions to give [Cu(SbF ₆)(BPEP)] and [Cu₂(BPEP)₂(μ-PF ₆)]⁺, respectively [BPEP = 2,6-bis(1-phenyl-2- phosphaethenyl)pyridine]. NMR and theoretical studies indicate a reduced anionic charge of the μ-PF₆ ligand, which is induced by the strong π-accepting ability of BPEP.

Full text of DOI:10.1039/c1cc11315a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 6332–6334
www.rsc.org/chemcomm
COMMUNICATION
Synthesis and coordination behavior of CuI bis(phosphaethenyl)pyridine
complexesw
Yumiko Nakajima,*ab Yu Shiraishi,b Takahiro Tsuchimotob and Fumiyuki Ozawa*ab
Received 7th March 2011, Accepted 12th April 2011
DOI: 10.1039/c1cc11315a
CuI complexes bearing BPEP as
phosphaalkene ligand undergo effective bonding interactions
a PNP-pincer type
ꢀ
ꢀ
with SbF6 and PF6 as non-coordinating anions to give
[Cu(SbF6)(BPEP)] and [Cu2(BPEP)2(l-PF6)]+, respectively
[BPEP = 2,6-bis(1-phenyl-2-phosphaethenyl)pyridine]. NMR
and theoretical studies indicate a reduced anionic charge of the
l-PF6 ligand, which is induced by the strong p-accepting ability
of BPEP.
Phosphaalkenes with PQC bond(s) are low-coordinate
phosphorus compounds that possess extremely low-lying p*
orbitals around the phosphorus atom, and thus exhibit strong
p-acceptor properties towards transition metals.1,2 We have
demonstrated with bidentate diphosphinidenecyclobutene
ligands (DPCB-Y) that this property is useful for catalysis,
leading to highly efficient organic transformations.3 More
recently, we have developed 2,6-bis[1-phenyl-2-(2,4,6-tri-tert-
butylphenyl)-2-phosphaethenyl]pyridine (BPEP) as a tridentate
PNP-pincer type phosphaalkene ligand, which successfully
stabilizes a coordinatively unsaturated 15-electron complex
with a high-spin FeI center.4
Scheme 1 Mes* = 2,4,6-tri-tert-butylphenyl.
complex 1 has the four-coordinate structure with a distorted
tetrahedral configuration as confirmed by X-ray diffraction
analysis (see ESIw). A similar structure has been found for the
complex with a PNP-pincer type phosphinine ligand, and
attributed to lower s-donating ability of low-coordinate
phosphorus ligands than phosphine ligands.7
This communication reports the synthesis and structures of
BPEP complexes of copper. Although PNP-pincer type
ligands are ongoing research topics in coordination chemistry,5
their copper complexes have been extremely limited.6–8 In this
study, we found that BPEP forms a CuI complex of a highly
Complex 1 reacted with silver salts of non-coordinating anions
(AgX; X = SbF6 and PF6) to afford complexes of the formula
‘‘CuX(BPEP)’’ [X = SbF6 (2a), PF6 (2b)]. Since both complexes
exhibited the same 31P NMR chemical shift (d 213.3) in CD2Cl2,
they should exist in ionic form in a polar solvent without direct
interaction between [Cu(BPEP)]+ and Xꢀ.
electron-deficient nature, causing effective bonding inter-
actions with ‘‘non-coordinating’’ anions such as SbF6ꢀ and PF6
ꢀ
.
Complex 2a readily coordinated with MeCN, CO, anꢀd
The BPEP ligand was introduced to CuBr in toluene at
90 1C (Scheme 1). The resulting [CuBr(BPEP)] (1) was
characterized by NMR spectroscopy and elemental analysis.
While it has been shown that [CuBr(pnp)] [pnp = 2,6-bis(di-
tert-butylphosphinomethyl)pyridine] as a phosphine analogue
adopts a three-coordinate structure without Cu–N bonding,6
tBuNC in CD2Cl2 to form [Cu(L)(BPEP)]+SbF6
t
[L = MeCN (3), CO (4), BuNC (5)], respectively. The n(CO)
band of 3 appeared at 2132 cmꢀ1 in the IR spectrum; the value is
close to that of free CO (2143 cmꢀ1). Complex 5 exhibited the
n(NC) band at 2198 cmꢀ1. This value is higherꢀthan that of the
phosphine analogue [Cu(tBuNC)(pnp)]+SbF6 (2177 cmꢀ1).6
These data illustrate the highly electron-deficient nature of the
copper center.
a International Research Center for Elements Science (IRCELS),
Institute for Chemical Research, Kyoto University, Uji,
Kyoto 611-0011, Japan. E-mail: nakajima@scl.kyoto-u.ac.jp,
ozawa@scl.kyoto-u.ac.jp
While the 31P NMR signal of 2a appeared at d 213.3 in
CD2Cl2, the signal was shifted downfield (d 234.2) in C6D6.
This tendency was quite different from that observed for 1,
which showed the same chemical shift in CD2Cl2 and C6D6
(d 251.5). As we have documented for DPCB-Y complexes,9
the 31P NMR chemical shift of the phosphaalkene ligand is
rather sensitive to the M–P bond length, and tends to increase
b Department of Energy and Hydrocarbon Chemistry,
Graduate School of Engineering, Kyoto University,
Uji, Kyoto 611-0011, Japan
w Electronic supplementary information (ESI) available: Experimental
procedures and analytical data, and computational details. CCDC
812261–812265. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c1cc11315a
c
6332 Chem. Commun., 2011, 47, 6332–6334
This journal is The Royal Society of Chemistry 2011

View Article Online
Fig. 1 ORTEP drawing of 2a with 50% probability ellipsoids.
Hydrogen atoms, disordered tert-butyl groups and the disordered F6
atom were omitted for clarity. Selected bond distances (A) and angles (1):
Cu–N 2.107(3), Cu–P1 2.2526(11), Cu–P2 2.2612(11), Cu–F1 2.190(3),
Sb–F1 1.916(3), Sb–F2 1.843(4), Sb–F3 1.859(5), Sb–F4 1.846(4),
Sb–F5 1.850(5), N–Cu–F1 146.05(15), P1–Cu–P2 156.25(5),
Cu–F1–Sb 143.9(2).
Fig. 2 ORTEP drawing of 6 with 50% probability ellipsoids.
Hydrogen atoms and the counter anion (PF6ꢀ) are omitted for clarity.
Selected bond distances (A) and angles (1): Cu–N 2.097(3), Cu–P1
2.2613(10), Cu–P2 2.2638(9), Cu–F1 2.241(2), P3–F1 1.637(2), P3–F2
1.586(2), P3–F3 1.588(2), N–Cu–F1 144.32(9), P1–Cu–P2 157.34(4),
N–Cu–F1 144.32(9), Cu–F1–P3 128.95(13).
as the M–P bond is elongated. Although there is the possibility
that the Cu–P distance of 2a varies with coordinating solvents,
it seems more likely that the SbF6ꢀ anion is associated with the
cationic [Cu(BPEP)]+ moiety to form a neutral complex in
nonpolar C6D6. To examine this point, single crystals of 2a
were grown from a toluene solution (87%), and examined by
X-ray diffraction analysis.
Scheme 2
Fig. 1 shows the structure of 2a. The copper center has a
distorted tetrahedral configuration with the N–Cu–F1 angle of
146.04(15)1 and the P1–Cu–P2 angle of 156.25(5)1. The
Cu–F1–Sb bond angle is 143.9(2)1, and the Sb atom adopts
a typical octahedral configuration. The Cu–N and Cu–P
lengths are in the range of PNP-pincer complexes (2.06–2.16
and 2.22–2.32 A, respectively).6–8
The most striking feature of 2a is the significantly short
Cu–F1 bond [2.190(3) A], which is comparable to that of CuI
fluorides (B2.1 A).10 Reflecting the occurrence of an effective
bonding interaction between the Cu and F1 atoms, the Sb–F1
bond is elongated by 0.057–0.073 A, compared with the other
Sb–F bonds. To the best of our knowledge, this is the first
example of a CuI complex with a coordinated SbF6 anion.
ꢀ
Next, we attempted to crystallize the PF6 complex 2b from a
toluene solution. However, the crystalline product obtained
was not 2b, but a cationic PF6-bridged dimer of Cu(BPEP)
units, having a PF6ꢀ counter anion (6). Fig. 2 shows the X-ray
structure of the cationic portion, which adopts Ci symmetry
with the P3 atom at the point of symmetry. Two Cu(BPEP)
units are connected by a m-PF6 group in a zigzag conformation.
The length of the Cu–F1 bond is 2.241(2) A, which is
somewhat longer than that of 2a, but still in the range of an
effective bonding interaction between the Cu and F atoms.11
Moreover, the P3–F1 bond [1.637(2) A] is clearly longer than
the other P–F bonds [1.586(2) and 1.588(2) A].
Fig. 3 (a) 19F and (b) 31P{1H} NMR spectra of 6 in CD2Cl2 at 20 1C.
identical to those of Bu4NPF6, this signal is assigned to PF6ꢀ as
the counter anion. Accordingly, the latter arises from the m-PF6
group. A remarkable feature of the spectrum is the significantly
1
weak intensity of the latter signal at d ꢀ77.8. The JPF value
ꢀ
(978 Hz) is clearly larger than that of the PF6 anion (711 Hz),
and comparable to that of neutral PF5 (938 Hz).12
The 31P{1H} NMR spectrum shown in Fig. 3(b) consists
of two sets of signals at d ꢀ17.3 (1JPF = 978 Hz) and ꢀ143.0
(1JPF = 711 Hz), arising from m-PF6 and PF6ꢀ, respectively.
The former signal appears as a triplet, while the latter shows
septet coupling as expected for PF6ꢀ; namely, the P–F
couplings for four out of the six fluorine atoms are missing from
the signal of the m-PF6 group. This phenomenon is rationalized by
assuming effective interactions of the four fluorine atoms with
copper centers having a quadrupole moment (I = 3/2) via the
rapid motion illustrated in Scheme 2. This behavior was not
frozen at ꢀ80 1C.
NMR spectroscopy revealed that the dimeric structure of 6
was preserved in CD2Cl2, even in the presence of excess MeCN
or CO. Furthermore, dynamic behavior on the NMR
time-scale was observed within the molecule (Scheme 2).
Fig. 3(a) shows the 19F NMR spectrum measured at room
temperature, showing two sets of doublets at d ꢀ73.7
(1JPF = 711 Hz) and ꢀ77.8 (1JPF = 978 Hz). Since the
1
chemical shift and the JPF constant for the former signal were
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6332–6334 6333

View Article Online
Fig. 4 The optimized structure and Mayer’s bond orders for a model
compound of 6, [(bpep)Cu–(m-PF6)–Cu(bpep)]+ (60).
To gain further insight into the bonding interaction between
copper and the m-PF6 group, DFT calculations were carried out.
Initially, we attempted geometry optimization for a model
compound of the cationic part of 6, [(bpep)Cu–(m-PF6)–
Cu(bpep)]+ (60), in which the 2,4,6-tri-tert-butylphenyl (Mes*)
and phenyl groups on the BPEP ligands were replaced by
hydrogen atoms. However, although the dimeric structure of 6
was reproduced, distances between the Cu and F atoms were
unreasonably shortened (Cu–F1 = 2.16 A; Cu–F2 = 2.48 A).
This is probably due to the absence of the bulky Mes* groups.
Therefore, the geometry of the Cu–(m-PF6)–Cu core was fixed to
the X-ray structure, and the remaining portion was optimized
assuming C2 symmetry around the F3–Cu–F3* axis.
Scheme 3
the high reactivity of 2a and 2b should be attributed to the high
electrophilicity of [Cu(BPEP)]+.
In summary, we have reported novel CuI complexes bearing
a phosphaalkene-based PNP-pincer ligand (BPEP). Thanks
to the strong p-accepting ability of the PQC bonds, the
[Cu(BPEP)]+ species possesses a highly electron-deficient
ꢀ
copper cꢀenter, exhibiting strong affinity towards SbF6
Fig. 4 shows the optimized geometry of 60 under the above
structural constraints. The Mayer’s bond orders (B) for Cu–F
and P–F bonds are also presented. There is evidence for a
bonding interaction between Cu and F1 (B = 0.32) and a
weakening of the P3–F1 bond (B = 0.66). It is also observed
that the F2 atom interacts with the Cu atom with a bond order
of 0.13, despite the long distance between those atoms
(3.054 A). Bonding interactions of F atoms with Cu centers
are also observed in several molecular orbitals (see ESIw).
Table 1 compares the charge distributions in [(bpep)Cu–
(m-PF6)–Cu(bpep)]+ (60) and [Cu(bpep)]+ (20), which were
evaluated by natural population analysis. The m-PF6 group of
60 is charged to ꢀ0.84, meaning that the negative charge of the
ꢀ
and PF6 as non-coordinating anions. Thus, SbF6 is
coordinated with [Cu(BPEP)]+ to form [Cu(SbF6)(BPEP)]
ꢀ
(2a) as a neutral species. On the other hand, PF6 formed
[Cu2(BPEP)2(m-PF6)]+PF6ꢀ (6). The dinuclear structure of 6 is
stable in solution even in the presence of excess MeCN or CO.
This is due to the occurrence of effective bonding interactions
between the Cu and F atoms.
This work was supported by a Grant-in-Aid for Scientific
Research from MEXT Japan and the JST PRESTO program.
Notes and references
1 P. Le Floch, Coord. Chem. Rev., 2006, 250, 627; F. Mathey, Angew.
Chem., Int. Ed., 2003, 42, 1578; L. Weber, Coord. Chem. Rev.,
2005, 249, 741; C. Muller and D. Vogt, Dalton Trans., 2007, 5505.
2 J. Dugal-Tessier, G. R. Dake and D. P. Gates, Org. Lett., 2010, 12,
4667; J. Dugal-Tessier, G. R. Dake and D. P. Gates, Organo-
metallics, 2007, 26, 6481.
3 F. Ozawa and M. Yoshifuji, Dalton Trans., 2006, 4987.
4 Y. Nakajima, Y. Nakao, S. Sakaki, Y. Tamada, T. Ono and
F. Ozawa, J. Am. Chem. Soc., 2010, 132, 9934.
ꢀ
¨
PF6 anion (ꢀ1.00) is reduced by bridging coordination with
two molecules of 20. Since the copper center of 60 is more
positively charged than thꢀat of 20, it is concluded that the
negative charge of the PF6 anion is distributed to the bpep
ligand upon the formation of 60, very probably via p-back-
bonding between copper and bpep.
5 J. I. van der Vlugt and J. N. H. Reek, Angew. Chem., Int. Ed., 2009,
48, 8832.
Complexes 2a and 2b undergo ionic dissociation in CD2Cl2
as a polar solvent (vide supra). It was found that the
complexes cleave the Si–N bond of Me3SiN3 to afford
[Cu2(BPEP)2(m-N3)]+Xꢀ [X = SbF6 (7a), PF6 (7b)] in 96
and 36% yields, respectively, along with Me3SiF (Scheme 3).
The reactions probably proceed via cooperative activation of
Me3SiN3 by the electrophilic copper center and the nucleophilic
fluoride ion. Since it is known that it is very difficult to dissociate a
fluoride ion from SbF6ꢀ and PF6ꢀ as non-coordinating anions,13
6 J. I. van der Vlugt, E. A. Pidko, D. Vogt, M. Lutz, A. L. Spek and
A. Meetsma, Inorg. Chem., 2008, 47, 4442; J. I. van der Vlugt,
E. A. Pidko, D. Vogt, M. Lutz and A. L. Spek, Inorg. Chem., 2009,
48, 7513; J. I. van der Vlugt, E. A. Pidko, R. C. Bauer,
Y. Gloaguen, M. K. Rong and M. Lutz, Chem.–Eur. J., 2011,
17, 3850.
7 C. Muller, E. A. Pidko, M. Lutz, A. L. Spek and D. Vogt,
¨
Chem.–Eur. J., 2008, 14, 8803.
8 A. Hayashi, M. Okazaki, F. Ozawa and R. Tanaka, Organo-
metallics, 2007, 26, 5246; A. Orthaber, F. Belaj and R. Pietshnig,
Inorg. Chim. Acta, 2011, DOI: 10.1016/j.ica.2011.02.078.
9 Y. Nakajima, M. Nakatani, K. Hayashi, Y. Shiraishi, R. Takita,
M. Okazaki and F. Ozawa, New J. Chem., 2010, 34, 1713.
10 P. C. Healy, J. V. Hanna, J. D. Kildea, B. W. Skelton and
A. H. White, Aust. J. Chem., 1991, 44, 427; B. F. Straub,
F. Rominger and P. Hofmann, Inorg. Chem., 2000, 39, 2113.
Table 1 Charge distribution in 60 and [Cu(bpep)]+ (20)
Complex
11 Weak coordination of PF6 with CuI at a distance of 2.609(2) ꢀA
ꢀ
[Cu2(bpep)2(m-PF6)]+ (60)
[Cu(bpep)]+ (20)
Component
has been reported for [Cu(p-CH2CQCHCOOCH3)(bpy)]+PF6
T. Pintauer, J. Organomet. Chem., 2006, 691, 3948.
:
Cu
bpep
m-PF6
+0.78
+0.14
ꢀ0.84
+0.67
+0.33
—
12 H. S. Gutowsky, D. W. McCall and C. P. Slichter, J. Chem. Phys.,
1953, 21, 279.
13 J. Devynck, A. B. Hadid, P. L. Fabre and B. Tremillon, Anal.
´
The values were determined by DFT calculations and NBO analysis.
Chim. Acta, 1978, 100, 343.
c
6334 Chem. Commun., 2011, 47, 6332–6334
This journal is The Royal Society of Chemistry 2011