Lithium cyanocuprates: Structural studies of a phosphane adduct of a lower-order lithium cyanocuprate

Article abstract of DOI:10.1002/ejic.200400685

Addition of PPh₃ to the lower-order cyanocuprate PhCuCNLi results in a significant increase in reactivity in 1,4-Michael addition reactions. This is explained by a marked structural change in the cyanocuprate reagent, observed in both THF solution and the solid state, to give [Ph ₂Cu]^- cuprate anions and a [Cu₂Li ₄(CN)₄(PPh₃)₄-(THF) ₁₀]²⁺ dication. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejic.200400685

SHORT COMMUNICATION
Lithium Cyanocuprates: Structural Studies of a Phosphane Adduct of a
‘‘Lower-Order’’ Lithium Cyanocuprate
Robert P. Davies*[a] and Stefan Hornauer[a]
Keywords: Cuprates / Lithium / Reaction intermediates / Michael addition
Addition of PPh3 to the ‘‘lower-order’’ cyanocuprate
PhCuCNLi results in a significant increase in reactivity in
give [Ph2Cu]− cuprate anions and a [Cu2Li4(CN)4(PPh3)4-
2+
(THF)10
]
dication.
1,4-Michael addition reactions. This is explained by
a
marked structural change in the cyanocuprate reagent, ob- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
served in both THF solution and the solid state, to
Germany, 2005)
Introduction
ity[15,16] studies of these higher-order cuprates have been less
conclusive predicting several possible solution structures.
However, the current consensus as laid out by Krause in a
recent review,[17] is that the reactive solution species consists
of a cuprate anion [R2Cu]Ϫ and a [LiϪCNϪLi]ϩ cation,
either as a solvent-separated ion pair (SSIP), such as 3, and
similar to the structures observed in the solid-state charac-
terizations, or possibly as some form of contact-ion pair
(CIP).
Lithium cyanocuprates are commonly used in the re-
gioselective formation of new carbonϪcarbon bonds, and
have been shown to have significant advantages for certain
reactions over classical Gilman reagents prepared from cop-
per(i) halides.[1,2] Cyanocuprates can conveniently be di-
vided into two distinct groups based upon their compo-
sition and reactivity: ‘‘lower order’’ cyanocuprates formed
from the 1:1 reaction of CuCN and LiR (R ϭ organo
group) to give species of the general formula RCu(CN)Li;
and ‘‘higher order’’ species from the 1:2 reaction of CuCN
and LiR and hence with the formulation R2Cu(CN)Li2.
Spectroscopic studies of lower-order cyanocuprates both in
solution[3] and the solid state[4Ϫ7] have shown 1, or aggre-
gates of 1 such as 2, to be the most likely structural motif
for these species (Scheme 1). Hence [(Me2PhSi)3C-
Cu(CN)Li(THF)3][4] is a monomer (1) in the solid state,
Scheme 1. Solid-state structural motifs for lower-order (1, 2) and
higher-order (3) lithium cyanocuprates
whilst [(Me2PhSi)3CCu(CN)Li(THF)2]2,[4] [(C6H3Ϫ2,6-
[5]
Trip2)Cu(CN)Li(THF)2]2
(Trip ϭ ϪC6H2-2,4,6-iPr3),
and [tBuCu(CN)Li-
[Me3SiCH2CuCNLi(OEt2)2]2,[6]
Despite these recent advances, there are still many gaps
in our knowledge of cyanocuprates, and this is particularly
true when we consider the use of additives. Boron trifluor-
ide, alkali-metal halides and phosphanes amongst others
have all been employed as additives for lithium cyanocup-
rates in order to enhance the stability, selectivity and/or re-
activity of these species.[1] Despite their reported success in
this field, little is know about their mode of operation and
the effect they have upon the molecular structure of the
organocopper reagents. This paper reports on the first
structural characterization of a phosphane adduct of a lith-
ium cyanocuprate and discusses its reactivity in Michael ad-
dition reactions.
[7]
(Et2O)2]ϱ
are all dimers (2). However, the structure of
higher-order cyanocuprates has been the subject of much
more controversy with arguments centering on whether the
cyanide group remains bound to the copper or not. Again
recent crystallographic studies have played a major part in
this discussion with structural characterization of the higher
order
cyanocuprates
and
[tBu2Cu{Li(THF)-
[(2-C6H4CH2NMe2)2Cu{Li-
(pmdeta)}2CN][7]
[8]
(THF)2}2CN]ϱ
both revealing ionic compounds con-
sisting of [R2Cu]Ϫ cuprate anions and [LiϪCNϪLi]ϩ cat-
ions (3). In contrast, infrared spectroscopic,[9] NMR
spectroscopic,[10Ϫ13] cryoscopic,[14] and kinetic reactiv-
[a]
Results and Discussion
Addition of triphenylphosphane to a THF/toluene solu-
tion of the lower-order cyanocuprate PhCuCNLi (formed
Dept. of Chemistry, Imperial College London,
South Kensington, London, SW7 2AZ, UK
Fax: (internat.) ϩ44-870-1300438
E-mail: r.davies@imperial.ac.uk
Eur. J. Inorg. Chem. 2005, 51Ϫ54
DOI: 10.1002/ejic.200400685
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
51

R. P. Davies, S. Hornauer
SHORT COMMUNICATION
from the 1:1 reaction of CuCN and LiPh) at 0 °C gave a unit of 4: Li(1) bridges two cyanide groups at their nitrogen
˚
˚
clear yellow solution from which crystals of 4 were grown ends [Li(1)ϪN(1) 2.036(12) A; Li(1)ϪN(1A) 2.095(12) A]
at Ϫ30 °C (Scheme 2). X-ray structural studies show 4 to and is solvated by two THF molecules [mean LiϪO, 1.919
˚
be an ionic species comprising
a
centrosymmetric A], whereas Li(2) is terminally bound to just one cyanide
[Cu2Li4(CN)4(PPh3)4(THF)10]2ϩ dication and two sym- group [Li(2)ϪN(2) 1.90(4) A] and has its quadruplet com-
˚
metry-related [CuPh2]Ϫ cuprate anions (Figure 1).
pleted by three THF molecules [mean LiϪO, 1.87 A]. In
˚
keeping
with
all
previously
reported
lithium
cyanocuprates,[4Ϫ7] the cyanide carbon atoms in 4 are essen-
tially linear [Cu(1)ϪC(9)ϪN(1), 175.5(6); Cu(1)ϪC(46)Ϫ
N(2), 176.2(8)°] and the CN unit is bent out of the plane
of the central Li2N2 ring (by 12.6°).
The [Ph2Cu]Ϫ cuprate anions in 4 contain close to linear
CuI centers [C(59)ϪCu(2)ϪC(65), 174.8(4)°] with CϪCu
˚
bond lengths of 1.900(11) A [Cu(2)ϪC(59)] and 1.931(11)
˚
A [Cu(2)ϪC(65)]. The dihedral angle between the phenyl
rings is 59.1°. These units are therefore structurally very
similar to the [R2Cu]Ϫ cuprate anions observed in the
higher-order
cyanocuprates
[tBu2CuLi2(CN)(THF)2-
(pmdeta)2][7] and [(2-C6H4CH2NMe2)2Cu{Li(THF)2}2-
CN]ϱ.[8]
Low temperature (Ϫ100 °C) 1H-decoupled 13C NMR
spectroscopy carried out on an aliquot of the CuCN ϩ
PhLi ϩ PPh3 reaction mix after stirring for 10 min at 0 °C
in toluene/THF solution shows singlets at δ ϭ 174.20 and
144.27 ppm for the phenyl ipso carbon and the cyanide car-
bon resonances, respectively. The chemical shift of the ipso
carbon is therefore downfield from the value observed for
Scheme 2. Synthesis of 4
PhCuCNLi in the absence of phosphane (δ
ϭ
166.05 ppm),[3] indicating some structural change has oc-
curred. Bertz has previously reported the chemical shifts
of the ipso carbon atoms of the higher-order cyanocuprate
formed from CuCN ϩ2 PhLi and also of the Gilman re-
agent from CuI ϩ2 PhLi as 174.33 and 174.13 ppm, respec-
tively (in [D8]THF at Ϫ78 °C).[10] Based on these almost
identical NMR shifts, he was able to argue that [CuPh2]Ϫ
cuprate subunits were present in each case. The fact that we
observe the ipso carbon shift to be of an almost identical
value to those observed by Bertz confirms that di-
phenylcuprate anions are present in solution for 4 as well
as in the solid state. The cyanide 13C NMR resonance (δ ϭ
144.27 ppm) differs from that observed for PhCuCNLi in
the absence of PPh3 (δ ϭ148.50 ppm)[3] and is significantly
Figure 1. Molecular structure of 4; only one of the two symmetry-
related [CuPh2]Ϫ cuprate anions is shown for clarity
The dication in 4 is unprecedented, containing both Liϩ upfield from the [Li2CN]ϩ cation cyanide resonance re-
and CuI ions linked together through cyanide groups, and ported for higher-order cyanocuprates (mean,
δ ϭ
can perhaps best be thought of as two PPh3-complexed 158.80 ppm).[10] However, the resonance is close to the
[Cu(CN)2]Ϫ anions aggregated with four THF-solvated Liϩ range of reported values for lower-order cyanocuprates (δ ϭ
cations. The copper atoms in the dication are approximately 45.51Ϫ151.20 ppm),[3Ϫ7] suggesting that ‘‘CuCNLi’’ units,
tetrahedral, and bond to two PPh3 groups [Cu(1)ϪP(1) similar or identical to those observed in the cation of 4, are
˚
˚
2.294(2) A; Cu(1)ϪP(2) 2.299(2) A] and to the carbon present in solution.
˚
atoms of two cyanide groups [Cu(1)ϪC(9) 1.944(6) A;
Infrared spectroscopic characterization of the addition
˚
Cu(1)ϪC(46) 1.952(9) A]. These CuϪCN distances are product of PhLi and CuCN (PhCuCNLi) gave a cyanide
therefore longer than those observed in previously reported stretch in THF solution at 2137 cmϪ1 (cf. 2133 cmϪ1 pre-
[4Ϫ7]
˚
lower-order cyanocuprates (mean, 1.906 A),
but are viously reported for MeCuCNLi).[9] On addition of PPh3,
˚
similar to the CuϪCN distances of 1.964(4) and 1.948(5) A this peak decreases significantly in size and a new peak is
reported for [Cu2(CN)2(PPh3)4(hppH)] (hppH
ϭ
observed at 2120 cmϪ1. This new absorption differs in fre-
1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]pyrimidine), quency from the cyanide stretch reported for the higher-
which also contains a CuI(CN)2(PPh3)2 metal center.[18] order cyanocuprate formed from CuCN ϩ 2MeLi (2115
Two different lithium ions are observed in the asymmetric cmϪ1), attributable to [LiϪCNϪLi]ϩ,[9] and instead can be
52
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 51Ϫ54

Structural Studies of a Phosphane Adduct of Lithium Cyanocuprates
SHORT COMMUNICATION
assigned to a cyanide stretch in a [Cu(PPh3)2CNLi]ϩ unit, 4 likely to favor the formation of SSIPs rather than the
which is also in agreement with the 13C NMR spectroscopic more reactive CIP species.[16]
and X-ray crystallographic observations (vide supra).
31P NMR spectroscopy on the reaction mixture in THF
containing excess triphenylphosphane (CuCN ϩ PhLi ϩ1.5 Conclusion
PPh3) reveals one peak (δ ϭ Ϫ2.54 ppm) at room tempera-
Addition of PPh3 to the ‘‘lower-order’’ cyanocuprate
PhCuCNLi drastically alters its structure by promoting the
formation of [CuPh2]Ϫ cuprate units both in THF solution
(observed using multi-nuclear and variable temperature
NMR spectroscopy) and in the solid state (observed using
single-crystal X-ray diffraction). Such cuprate units have
been previously reported in ‘‘higher-order’’ cyanocuprates
and Gilman cuprates, but this is this first time they have
been observed in a formally ‘‘lower-order’’ cyanocuprate.
The addition of PPh3 is also shown to increase the reactiv-
ity of PhCuCNLi in Michael addition reactions with 2-
cyclohexanone, and this increase in reactivity can be ex-
plained by the observed structural changes in the reagent
on addition of phosphane, in particular the generation of
the anionic [R2Cu]Ϫ cuprate units.
ture. On cooling to Ϫ100 °C, this peak splits into two new
resonances: a broad peak at δ ϭ 0.04 ppm and a sharp peak
at δ ϭ Ϫ7.76 ppm. These resonances can be assigned to
[Cu(CN)2]Ϫ-complexed PPh3 and free PPh3, respectively,
and interchange of the two types of PPh3 occurs at room
temperature on the NMR time scale. 7Li NMR spec-
troscopy on the same reaction mixture reveals two reson-
ances (δ ϭ Ϫ2.44 and Ϫ3.37 ppm), consistent with the
bridging and terminal lithium atoms present in the cation
of 4.
The addition of the phosphane therefore results in a con-
siderably different structural motif for 4 relative to all pre-
viously reported lower-order cyanocuprates (1, 2), and the
most significant difference, in terms of its effect on the reac-
tivity, undoubtedly is the generation of [Ph2Cu]Ϫ cuprate
anions. The observed increased reactivity of higher- than
lower order cyanocuprates has previously been attributed
to the presence of such [R2Cu]Ϫ cuprate units,[17] and it
seems likely that any increase in reactivity for PhCuCNLi
on addition of PPh3 is due to the generation of these anions.
However, although it has previously been suggested that ad-
dition of PPh3 to cuprates does indeed increase their reac-
tivity,[1] no systematic studies have been reported. We there-
fore obtained logarithmic reactivity profiles (LRPs)[15] to
compare the rates of reaction of (a) PhCuCNLi, (b)
Experimental Section
General Remarks: CuCN (Fluka Ͼ 99%), PPh3 (Aldrich 99%) and
2-cyclohexene-1-one (Lancaster Synthesis 97%) were used without
further purification. Pure, halogen-free PhLi was synthesized prior
to use by reaction of equimolar amounts of PhI with nBuLi in
hexane at 0 °C.[19] THF and diethyl ether were distilled from over
sodium wire/benzophenone. Unless otherwise stated, all manipu-
lations were carried out under nitrogen using either a glove box or
PhCuCNLi ϩ PPh3 (4), and (c) Ph2CuCNLi2 in Michael double-manifold vacuum line. Dry ice/2-propanol baths were used
for reactions at Ϫ78 °C. GC analyses were performed with an Agil-
ent GC4890 gas chromatograph equipped with FID detector and
a 25-m SGE BPX5 capillary column [0.22 mm i.d., 0.25 micron
film, cross-linked 5% phenyl (equiv.) polysilphenylene-siloxane].
The GC was calibrated with authentic product (3-phenylcyclohexa-
none) and dodecane (Aldrich 99%) as internal standard. NMR
spectra were recorded on a Jeol EX270 Delta Upgrade or a Bruker
AM-500 spectrometer. External standards used were TMS (1H,
13C) 85% H3PO4 (31P) and LiCl/D2O (7Li).
addition reactions with 2-cyclohexanone (Table 1). The
LRP was generated by quenching the reaction after a series
of times spanning several orders of magnitude, and measur-
ing the yields of 3-phenylcyclohexanone using GLC and the
internal standard method.[15] The lower order cyanocuprate
PhCuCNLi is very unreactive under these conditions with
no product formation observed even after 1 h. However, ad-
dition of PPh3 increases the yields significantly to 29% after
1 h. The yields are larger still for the higher order cyanoc-
uprate and this can be explained by double the concen-
tration of cuprate anions in solution for Ph2CuCNLi2 than
4, and also by the differing nature of the cations in the two
reagents, with the larger more sterically hindered cations in
Synthesis of [Cu2Li4(CN)4(PPh3)4(THF)10]·2[Ph2Cu] (4): A solution
of PhLi (1.8 m in cyclohexane/diethyl ether; 1.1 mL, 2.0 mmol) was
added to a suspension of CuCN (0.180 g; 2.0 mmol) in THF
(10 mL) at Ϫ78 °C followed by dropwise addition of PPh3 (0.577 g;
2.2 mmol) in THF (5 mL). The resulting clear yellow solution was
warmed to 0 °C, filtered through celite and reduced in vacuo to
10 mL. On storage for 72 h at Ϫ30 °C, crystals of 4 were obtained
in 46% yield (0.57 g). M.p. 118Ϫ122 °C (dec.). Note that NMR
studies on the reaction mix, as detailed in the main text, indicate a
near quantitative yield based on PhLi and CuCN consumption.
Due to the extreme air and moisture sensitive nature of 4 we were
unable to obtain satisfactory elemental analyses. 1H NMR spec-
troscopy (270 MHz, [D6]DMSO, 298 K): δ ϭ 7.09Ϫ7.44 (m, 70 H,
PPh3, CuPh2), 3.58 (m, 20 H, THF), 1.72 (m, 20 H, THF) ppm Ϫ
note that some THF was lost during the isolation procedure. 13C
NMR spectroscopy (101 MHz, THF, 173 K): δ ϭ 174.20 (CN),
141.33 (Cu-ipso-Ph), 137.49Ϫ124.97 (Cu-o/m/p-Ph, PPh3), 67.57
(THF), 25.84 (THF) ppm. 31P NMR spectroscopy (109.38 MHz,
Table 1. LRP data for reactions of (a) PhCuCNLi, (b) 4 and (c)
Ph2CuCNLi2 with 2-cyclohexanone in THF, showing the yields (%)
of 3-phenylcyclohexanone
Reagent[a]
Time [h]
0.01
0.001
0.1
1
(a) PhCuCNLi
(b) PhCuCNLi ϩ PPh3 (4)
(c) Ph2CuCNLi2
0
1
19
0
3
53
0
22
66
0
29
70
[a] Reactions were carried out at Ϫ78 °C on a 1 mmol scale at 0.1 m
in THF. Quench: 6 mL of saturated aqueous ammonium chloride/
ammonia (90:10).
THF, 298 K):
δ
ϭ
Ϫ2.54 ppm. 7Li NMR spectroscopy
Eur. J. Inorg. Chem. 2005, 51Ϫ54
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 53

R. P. Davies, S. Hornauer
SHORT COMMUNICATION
(194.31 MHz, THF, 298 K): δ ϭ Ϫ2.44, Ϫ3.36 ppm. IR (THF, 25
°C): ν˜ ϭ 2120 (s) cmϪ1 (CϵN).
absorption-corrected reflections with F2 Ͼ 2σ(F2), S ϭ 1.035, 667
parameters. Residual electron density extrema Ϯ1.165 e·AϪ3. Since
˚
it proved impossible to differentiate between the cyanide carbon
and nitrogen atoms crystallographically, these atoms have been as-
signed based on previously reported structural characterizations of
cyanocuprates, all of which contain the cyanide unit attached to
the copper through the carbon end and linear carbon atoms.[4,7]
CCDC-226575 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge
CB2 1EZ, UK; Fax: (internat.): ϩ44-1223-336-033; E-mail:
deposit@ccdc.cam.ac.uk].
Logarithmic Reactivity Profiles (LRPs): The three cuprate reagents
used were prepared as follows: (a) PhCuCNLi. CuCN (4 mmol,
358 mg) and PhLi (4 mmol, 336 mg) were weighed out into a
250 mL Schlenk flask equipped with a stirrer bar, and cold THF
(40 mL, Ϫ78 °C) was added, and the flask was maintained at a
temperature of Ϫ78 °C. The reaction mixture was stirred for 6 min
and then warmed up to 0 °C for 6 min to anneal the reagent before
it was cooled back to Ϫ78 °C; (b) PhCuCNLi ϩ PPh3 (4). CuCN
(4 mmol, 358 mg) and PPh3 (4 mmol, 336 mg) were weighed out
into a 250 mL Schlenk flask equipped with a stirrer bar. PhLi
(4 mmol, 336 mg) was weighed out into a separate Schlenk flask
equipped with a stirrer bar. Cold THF (20 mL, Ϫ78 °C) was added
with a syringe into each Schlenk flask, and both solutions were
stirred for 10 min at room temperature making sure that the PhLi
had completely dissolved. Both Schlenk flasks were then immersed
in a cooling bath (Ϫ78 °C), and the PhLi solution was transferred
with a cannula into the THF suspension of CuCN/PPh3. The reac-
tion mixture was stirred for 6 min and then warmed up to 0 °C for
6 min to anneal the reagent before it was cooled back to Ϫ78 °C;
(c) Ph2CuCNLi2. CuCN (4 mmol, 358 mg) and PhLi (8 mmol,
672 mg) were weighed out into a 250 mL Schlenk flask equipped
with a stirrer bar. Cold THF (40 mL, Ϫ78 °C) was added, and the
flask was maintained at a temperature of Ϫ78 °C. The resulting
cloudy suspension was stirred at Ϫ78 °C for 18 min to give a clear
yellow solution.
Acknowledgments
We thank the donors of the American Chemical Society Petroleum
Research Fund for support of this research.
[1] [1a]
R. J. K. Taylor, Organocopper Reagents (Series Eds.: L. M.
[1b]
Harwood, C. J. Moody), Oxford University Press, 1994.
J.
A. Kozlowski, in Comprehensive Organic Synthesis (Eds.: B. M.
Trost, I. Fleming), Pergamon, Oxford, 1991, vol. 4, pp.
169Ϫ198. [1c] B. H. Lipshutz, in Comprehensive Organometallic
Chemistry II (Eds.: E. W. Abel, F. G. A. Stone, G. Wilkinson),
Pergamon, Oxford, 1995, vol. 12, pp. 194Ϫ213. [1d]N. Krause,
A. Gerold, Angew. Chem. Int. Ed. Engl. 1997, 36, 186Ϫ204.
[1e]
N. Krause, Modern Organocopper Chemistry, Wiley-VCH,
For all cuprates (a)Ϫ(c): After a further 6 min of stirring at Ϫ78
°C, 10 mL of the reagent was transferred with a syringe into each
of four 25 mL flasks. The flasks were predried overnight in an oven
at 120 °C and taken into the glove box where they were equipped
with a stirrer bar and a rubber septum. They were kept under a
positive nitrogen pressure throughout the experiment. A solution
of 2-cyclohexene-1-one (96 mg, 1 mmol) in THF (1 mL) was added
to the flask with a syringe. The stopwatch was started when the
plunger was pushed. After the appropriate LRP time, a quench
solution consisting of 6 mL of saturated aqueous ammonium chlo-
ride/ammonia (90:10) was injected. The LRP times used were 1 h ϭ
60 min, 0.1 h ϭ 6 min, 0.01 h ϭ 36 s, 0.001 h ϭ 4 s. After quench-
ing the reaction, the flask was removed from the dry ice/2-propanol
bath and warmed in warm water until it reached room temperature.
The internal GC standard (dodecane, 227 µL, 1 mmol) was then
added. The aqueous layer was removed using a pipette, and the
organic layer filtered into a vial. The flask and residue were rinsed
with diethyl ether (5 mL), and the combined organic layer and
ether extract were dried over MgSO4. A sample of the quenched
reaction mixture was then run on the GC to ascertain the percent-
age yield of 3-phenylcyclohexanone. Repeat experiments indicated
an experimental error of up to a maximum of Ϯ10% (see reference
15 for a thorough discussion of experimental errors in LRP studies
of cuprates).
Weinheim, 2002.
[2]
[3]
[4]
B. H. Lipshutz, Synlett 1990, 119Ϫ128.
S. H. Bertz, J. Am. Chem. Soc. 1991, 113, 5470Ϫ5471.
C. Eaborn, M. S. Hill, P. B. Hitchcock, J. D. Smith, Organomet-
allics 2000, 19, 5780Ϫ5783.
C.-S. Hwang, P. P. Power, J. Am. Chem. Soc. 1998, 120,
6409Ϫ6410.
F. Bosold, M. Marsch, K. Harms, G. Boche, Z. Kristallogr.
New Cryst. Struct. 2001, 216, 143Ϫ144.
G. Boche, F. Bosold, M. Marsch, K. Harms, Angew. Chem. Int.
Ed. 1998, 37, 1684Ϫ1685.
C. M. P. Kronenburg, J. T. B. H. Jastzebski, A. L. Spek, G.
van Koten, J. Am. Chem. Soc. 1998, 120, 9688Ϫ9689.
H. Huang, K. Alvarez, Q. Lui, T. M. Branhart, J. P. Snyder, J.
E. Penner-Hahn, J. Am. Chem. Soc. 1996, 118, 8808Ϫ8816.
S. H. Bertz, J. Am. Chem. Soc. 1990, 112, 4031Ϫ4032.
B. H. Lipshutz, B. James, J. Org. Chem. 1994, 59, 7585Ϫ7587.
S. H. Bertz, G. Miao, M. Eriksson, Chem. Commun. 1996,
815Ϫ816.
X. Xie, C. Auel, W. Henze, R. M. Gschwind, J. Am. Chem.
Soc. 2003, 125, 1595Ϫ1601.
A. Gerold, J. T. B. H. Jastrzebski, C. M. P. Kronenburg, N.
Krause, G. van Koten, Angew. Chem. Int. Ed. Engl. 1997, 36,
755Ϫ757.
S. H. Bertz, A. Chopra, M. Eriksson, C. A. Ogle, P. Seagle,
Chem. Eur. J. 1999, 5, 2680Ϫ2691.
M. John, C. Auel, C. Behrens, M. Marsch, K. Harms, F.
Bosold, R. M. Gschwind, P. R. Rajamohanan, G. Boche,
Chem. Eur. J. 2000, 6, 3060Ϫ3068.
N. Krause, Angew. Chem. Int. Ed. 1999, 38, 79Ϫ81.
M. P. Coles, P. B. Hitchcock, Polyhedron 2001, 20, 3027Ϫ3032.
E. Wehman, J. T. B. H. Jastrzebski, J.-M. Ernsting, D. M.
Grove, G. van Koten, J. Organomet. Chem. 1988, 353,
133Ϫ143.
G. Sheldrick, SHELXTL PC version 5.1, Bruker AXS, Madi-
son, WI, 1997 and SHELX-97, Institut Anorg. Chemie, Tam-
mannstr. 4, 37077 Göttingen, Germany, 1998.
Received August 9, 2004
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
X-ray Crystallographic Data for 4: C147H168Cu4Li4N4O10P4, Mr ϭ
2556.65, monoclinic, space group P21/c, a ϭ 20.240(2), b ϭ
[17]
[18]
[19]
3
˚
˚
16.112(3), c ϭ 24.575(3) A, β ϭ 113.37(2)°, V ϭ 7356.6(18) A ,
Z ϭ 2 (Ci symmetry), ρcalcd. ϭ 1.154 g·cmϪ3, Mo-Kα radiation, λ ϭ
0.71069 A, µ ϭ 0.668 mmϪ1, T ϭ 230(2) K. 14981 unique data
˚
(Rint ϭ 0.1147, θ Ͻ 27.24°) were collected on a Nonius Kappa
CCD diffractometer. The structure was solved by direct methods
and refined by full-matrix least-squares on F2 values of all data
using the SHELXTL and SHELX-97 program systems,[20] to give
wR2 ϭ 0.3492, conventional R ϭ 0.1069 for 7011 observed
[20]
54
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 51Ϫ54