Copper(I) diphosphine catalysts for C-N bond formation: Synthesis, structure, and ligand effects

Article abstract of DOI:10.1021/om800139v

A series of copper(I) complexes containing bis(diarylphosphino)propane, bis(diarylphosphino)ethane, bis(diarylphosphino)methane, and N,N-bis(diarylphosphino)amine ligands (Aryl = Ph, 2-C₆H ₄(Me) or 2-C₆H₄(i-Pr)) has been synthesized. Crystal structures of selected chloride derivatives are reported. The complex structures proved to be very sensitive to both the backbone and P-substituents of the chelating ligand. The complexes have been screened for catalytic amidation reactions. Although in most cases only very low activity is observed, comparable with simple copper halide salts, notable exceptions are catalysts based on N,N-bis(diphenylphosphino)amine ligands, where a significant improvement in catalyst efficiency is observed. We propose the unusual electronic properties of these ligands may be the cause of their distinctive performance in these and other catalytic reactions where hard donor ligands have previously been employed.

Full text of DOI:10.1021/om800139v

3196
Organometallics 2008, 27, 3196–3202
Copper(I) Diphosphine Catalysts for C-N Bond Formation:
Synthesis, Structure, and Ligand Effects
Stephen Daly, Mairi F. Haddow, A. Guy Orpen, Giles T. A. Rolls, Duncan F. Wass,* and
Richard L. Wingad
School of Chemistry, UniVersity of Bristol, Cantock’s Close, Bristol, BS8 1TS, U.K.
ReceiVed February 15, 2008
A series of copper(I) complexes containing bis(diarylphosphino)propane, bis(diarylphosphino)ethane,
bis(diarylphosphino)methane, and N,N-bis(diarylphosphino)amine ligands (Aryl ) Ph, 2-C6H4(Me) or
2-C6H4(i-Pr)) has been synthesized. Crystal structures of selected chloride derivatives are reported. The
complex structures proved to be very sensitive to both the backbone and P-substituents of the chelating
ligand. The complexes have been screened for catalytic amidation reactions. Although in most cases
only very low activity is observed, comparable with simple copper halide salts, notable exceptions are
catalysts based on N,N-bis(diphenylphosphino)amine ligands, where a significant improvement in catalyst
efficiency is observed. We propose the unusual electronic properties of these ligands may be the cause
of their distinctive performance in these and other catalytic reactions where hard donor ligands have
previously been employed.
where softer donors are usually the ligand of choice. To our
knowledge no reports exist on the use of phosphines as
ligands in copper-catalyzed amidation, and this lack of results
for copper phosphine systems is maybe surprising given the
ubiquitous role of P-donor ligands in virtually all catalysis
with late transition metals. We have a long standing interest
in the application of small bite angle diphosphines as ligands
for catalysis and have previously shown that such ligands
can lead to active systems in, for example, olefin polymer-
ization with late transition metals, another area where hard
donors dominate.7 We describe here the synthesis of a range
of copper(I) diphosphine complexes and their screening for
the amidation of aryl halides.
1. Introduction
Palladium-catalyzed C-N bond formation has developed into
a versatile way of producing aryl amines and amides from aryl
halides,1 probably the best known methodology being the
Buchwald-Hartwig amination reaction, which makes use of
palladium complexes supported by strong σ-donor phosphine
or N-heterocyclic carbene (NHC) ligands.2 However, there has
also been recent progress in revisiting the much older Ullmann
and Goldberg copper-mediated methodologies,3,4 so that Buch-
wald and co-workers have demonstrated that efficient catalytic
systems operating at much lower temperatures than traditionally
employed with copper can be attained by judicious choice of
supporting ligand.5 These copper-based systems show advan-
tages over palladium catalysts with certain substrates, and the
two catalyst families are in many ways complementary.
2. Results and Discussion
A focused library of diphosphine ligands was targeted in
which the chelate length is systematically varied and an ortho
aryl substitutent added (Figure 1). The latter factor proved to
be crucial in previous catalysis with such ligands.7,8 We also
utilized the related N,N-bis(diarylphosphino)amines, which have
a track record of success in other reactions often associated with
harder donor sets.8
2.1. Synthesis of Copper Diphosphine Complexes: 13-24.
Copper diphosphine complexes with the general stoichiom-
etries [LCuCl]n (n ) 1-3), [L(CuCl)2], or [L2Cu]Cl were
synthesized in a straightforward manner by reaction of ligand
with CuCl in toluene (Scheme 1), with products being
obtained as white or off-white microcrystalline solids in
moderate to good yields (48-96%). One equivalent of
Bidentate chelating phosphines have been used as ligands
for copper-catalyzed amination of aryl halides with aryl
amines. However their activity is generally surpassed by
supporting ligands based on “hard” N and/or O donor
chelates. This is in contrast to palladium-catalyzed coupling,
6
* Corresponding author. E-mail: duncan.wass@bristol.ac.uk.
(1) Reviews: (a) Schlummer, B.; Scholz, U. AdV. Synth. Catal. 2004,
346, 1599. (b) Muci, A. R.; Buchwald, S. L. Top. Curr. Chem. 2002, 219,
131. (c) Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999, 576,
125. (d) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2046. (e) Wolfe,
J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem. Res. 1998,
31, 805.
(2) (a) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1348. (b) Louie, J.; Hartwig, J. F. Tetrahedron
Lett. 1995, 36, 3609.
(3) (a) Ullmann, F. Chem. Ber. 1903, 36, 2389. (b) Goldberg, I. Ber.
Dtsch. Chem. Ges. 1906, 39, 1691.
(6) Kelkar, A. A.; Patil, N. M.; Chaudhari, R. V. Tetrahedron Lett. 2002,
43, 7143.
(4) For recent reviews see:(a) Thomas, A. W.; Ley, S. V. Angew. Chem.,
Int. Ed. 2003, 42, 5400. (b) Beletskaya, I. P.; Cheprakov, A. V. Coord.
Chem. ReV. 2004, 248, 2337.
(7) Dennett, J. N. L.; Gillon, A. L.; Heslop, K.; Hyett, D. J.; Fleming,
J. S.; Lloyd-Jones, C. E.; Orpen, A. G.; Pringle, P. G.; Wass, D. F.; Scutt,
J. N.; Weatherhead, R. H. Organometallic. 2004, 23, 6077.
(8) (a) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.;
Wass, D. F. Chem. Commun. 2002, 858. (b) Cooley, N. A.; Green, S. M.;
Wass, D. F.; Heslop, K.; Orpen, A. G.; Pringle, P. G. Organometallics 2001,
20, 4769. (c) Dossett, S. J.; Gillon, A.; Orpen, A. G.; Fleming, J. S.; Pringle,
P. G.; Wass, D. F.; Jones, M. D. Chem. Commun. 2001, 699.
(5) (a) Klapers, A; Antilla, J. C.; Huang, X. H.; Buchwald, S. L.
J. Am. Chem. Soc. 2001, 123, 7727. (b) Klapars, A.; Huang, X.; Buchwald,
S. L. J. Am. Chem. Soc. 2002, 124, 7421. (c) Streiter, E. R.; Blackmond,
D. G.; Buchwald, S. L. J. Am. Chem. Soc. 2005, 127, 4120. (d) Shafir,
A.; Buchwald, S. L. J. Am. Chem. Soc. 2006, 128, 8742. (e) Altman,
R. A.; Buchwald, S. L. Nat. Protoc. 2007, 2, 2474.
10.1021/om800139v CCC: $40.75
2008 American Chemical Society
Publication on Web 05/30/2008

Copper(I) Diphosphine Catalysts
Organometallics, Vol. 27, No. 13, 2008 3197
Figure 1. Diphosphine ligand library.
Scheme 1. Complex Synthesis
Figure 2. Crystal structure of [(Ph2PCH2PPh2)CuCl]3 (13). Hy-
drogens and solvent are omitted for clarity.
Table 3. Selected Bond Lengths (Å) and Angles (deg) for Complex 14
Table 1. Selected Bond Lengths (Å) and Angles (deg) for Complex
13
Cu1-Cl1
Cu1-P1
2.2316(15)
2.2524(14)
2.2490(15)
2.2327(14)
2.2623(14)
2.2746(15)
118.87(6)
119.32(5)
Cu1-P1-C1-C2
-56.3(4)
178.4(4)
-175.4(4)
79.4(4)
Cu1-P1-C8-C9
Cu1-Cl1
Cu1-Cl2
Cu1-P1
Cu1-P3
Cu2-Cl1
Cu2-Cl2
Cu2-P2
Cu2-P5
Cu3-Cl2
Cu3-Cl3
2.4490(11)
2.3848(11)
2.2680(12)
2.2524(10)
2.4212(11)
2.4032(12)
2.2302(11)
2.2295(11)
2.4697(11)
2.3055(11)
Cu3-P4
2.2966(11)
2.2513(11)
74.28(4)
Cu1-P3
Cu2-P2-C16-C17
Cu2-P2-C23-C24
Cu1-P3-C30-C31
Cu1-P3-C37-C38
Cu2-P4-C45-C46
Cu2-P4-C52-C53
Cu3-P6
Cu2-Cl2
Cu2-P2
Cu2-P4
P1-Cu1-P3
P2-Cu2-P4
Cu1-Cl1-Cu2
Cu1-Cl2-Cu2
Cu1-Cl2-Cu3
Cu2-Cl2-Cu3
P1-C1-P2
P3-C26-P4
P5-C51-P6
71.6(4)
75.78(3)
-172.7(4)
174.4(4)
-59.1(4)
103.34(4)
105.82(4)
112.62(18)
116.02(17)
113.58(18)
Table 4. Selected Bond Lengths (Å) and Angles (deg) for Complex 17
Cu1-Cl1
Cu1-P1
2.1697(17) P1-C1-C2-P2
2.2471(16) Cu1-P1-C3-C4
2.2617(18) Cu1-P1-C10-C11
49.0(7)
23.25.4(6)
-67.0(5)
Table 2. Selected Bond Lengths (Å) and Angles (deg) for Complex
16
Cu1-P2
P1-Cu1-P2
C17A-P2-C24A 103.8(5)
92.98(6)
Cu1-P2-C17A-C18A -13.7(10)
Cu1-P2-C24A-C25A -69.0(5)
Cu1-Cl1
Cu1-Cl2
Cu2-Cl1
Cu2-Cl2
Cu3-Cl3
Cu1-P1
Cu1-P5
2.4027(19)
2.436(2)
2.409(2)
Cu2-P2
2.2463(19)
2.235(2)
2.214(2)
2.231(2)
-112.6(4)
172.4(4)
166.7(3)
Cu2-P3
Cu3-P4
Cu3-P6
P1-C13-C14-P2
P3-C39-C40-P4
P5-C65-C66-P6
trometry show this is the case. Ligand 3 does not react
smoothly with copper(I) chloride to give a [(3)CuCl]n
complex. After mixing ligand 3 with copper(I) chloride in
toluene at room temperature the 31P NMR spectrum shows a
broad peak at δ -22.0 indicating copper complexation, but
a large amount of unreacted ligand is also present (δ -49.5).
Heating this mixture at reflux for several days resulted in no
change to the product distribution, but the addition of a
further equivalent of copper(I) chloride led to the consump-
tion of the remaining free ligand. Elemental analysis and mass
spectrometry suggest a complex (15) with the composition
[(3)2Cu4Cl4] in the solid state (see Experimental Section).
1,3-Bis(diphenylphosphino)propane (7) also does not react
to give the complex [(7)CuCl]. In this case two broad
resonances are observed in the 31P NMR spectrum, in line
with the report of Pettinari et al., proposing this is due to an
equilibrium between [(7)2Cu2(µ-Cl)2] and [(7)2Cu]Cl.10 El-
emental analysis and electrospray mass spectrometry confirm
the presence of [(7)2Cu]Cl (19).
2.374(2)
2.2134(19)
2.2401(19)
2.2351(19)
diphosphine ligand per copper was used in each case, even
though this did not always lead to the expected stoichiometry
for the final product.
In general, elemental analyses data are consistent with the
formation of complexes of the type [LCuCl]n. It has been
reported that such complexes (specifically 13 and 16) are
not simple mononuclear species, the number of copper atoms
being sensitive to both ligand effects and solvent in solution.9
It therefore seems likely that higher nuclearity species are
formed; X-ray crystallography (see below) and mass spec-
(9) (a) Marsich, N.; Camus, A.; Cebulec, E. J. Inorg. Nucl. Chem. 1972,
34, 933. (b) Bera, J. K.; Nethaji, M.; Samuelson, A. G. Inorg. Chem. 1999,
38, 218. (c) Di Nicola, C.; Effendy; Fazaroh, F.; Pettinari, C.; Skelton, B. W.;
Somers, N.; White, A. H. Inorg. Chim. Acta 2005, 358, 720. (d) Cingolani,
A.; Di Nicola, C.; Effendy; Pettinari, C.; Skelton, B. W.; Somers, N.; White,
A. H. Inorg. Chim. Acta 2005, 358, 748. (e) Effendy; Di Nicola, C.; Pettinari,
C.; Pizzabiocca, A.; Skelton, B. W.; Somers, N.; White, A. H. Inorg. Chim.
Acta 2006, 359, 64.
(10) Effendy; Di Nicola, C.; Fianchini, M.; Pettinari, C.; Skelton, B. W.;
Somers, N.; White, A. H. Inorg. Chim. Acta 2005, 358, 763.

3198 Organometallics, Vol. 27, No. 13, 2008
Daly et al.
Table 5. Selected Bond Lengths (Å) and Angles (deg) for Complex 23
Cu1-Cu2
Cu1-Cu3
Cu2-Cu4
Cu3-Cu4
Cu1-Cl1
Cu1-Cl3
Cu2-Cl1
Cu2-Cl4
Cu2-Cl2
Cu3-Cl2
Cu3-Cl3
Cu4-Cl2
Cu4-Cl4
Cu1-P1
Cu2-P2
Cu3-P3
Cu4-P4
2.7750(13)
2.8699(10)
2.8726(9)
Cl1-Cu1-Cl3
105.44(4)
95.31(4)
91.22(5)
85.75(4)
105.57(4)
94.78(4)
121.60(14)
120.36(14)
42.9(3)
62.9(2)
65.4(3)
46.9(3)
58.5(3)
Cl1-Cu2-Cl4
Cl2-Cu2-Cl4
2.7526(13)
2.3045(10)
2.2932(14)
2.4982(12)
2.3393(10)
2.5579(13)
2.3414(12)
2.3355(10)
2.4490(10)
2.3085(13)
2.1756(10)
2.1920(10)
2.1866(10)
2.1867(9)
Cl1-Cu2-Cl2
Cl2-Cu3-Cl3
Cl2-Cu4-Cl4
P1-N1-P2
P3-N2-P4
Cu1-P1-C2-C3
Cu1-P1-C9-C10
Cu2-P2-C16-C17
Cu2-P2-C23-C24
Cu3-P3-C31-C32
Cu3-P3-C38-C39
Cu4-P4-C45-C46
Cu4-P4-C52-C53
44.3(3)
28.9(3)
75.4(3)
The reactivity patterns for bis(diarylphosphino)methylamine
ligands are similarly complex. The ortho tolyl derivative for
which we were able to obtain structural characterization (23) is
confirmed to have four copper chloride units and a ligand:Cu
ratio of 0.5 in the solid state. Other analysis is consistent with
this formulation. Electrospray mass spectrometry suggests other
derivatives to have trimeric or tetrameric structures. We have
had difficulty in obtaining satisfactory elemental analysis for
these derivatives (data presented in Experimental Section for
information), which we propose is due to contamination with
small amounts of substoichiometric ligand:Cu species.
Figure 4. Crystal structure of [(Ar2PCH2PAr2)CuCl]2 {Ar )
2-C6H4(Me)} (14) (hydrogens and solvent omitted for clarity).
Providing a general rationale for the observed reactivity and
structural patterns is challenging, as remarked in previous
studies.11 Less sterically encumbered, smaller bite-angle ligands
tend to form lower nuclearity species, although the interplay of
these factors is difficult to predict. Increasing steric bulk also
leads the copper atoms to adopt a more linear arrangement. The
influence of cuprophilic interactions also cannot be ruled out.12
It should also be noted that although these chloride complexes
are useful models in defining general trends, it is the amide
derivatives, formed under catalytic conditions, which are
believed to be the active species.
2.1.1. Crystal Structure of [(Ph2PCH2PPh2)CuCl]3 · 2C2H4-
Figure 5. Crystal structure of [(Ar2PCH2CH2PAr2)CuCl] {Ar )
2-C6H4(Me)} (17) (hydrogens omitted for clarity).
Cl2 (13). The dichloroethane solvate 13 crystallizes in the space
j
group P1 and has one molecule of complex and two of
Figure 6. Crystal structure of [(Ar2PN(Me)PAr2)Cu2Cl2]2 {Ar )
2-C6H4(Me)} (23) (hydrogens and solvent omitted for clarity).
dichloroethane in the asymmetric unit. The complex consists
of a triangle of copper(I) atoms, each edge of which is bridged
by a bis(diphenylphosphino)methane (dppm) ligand, creating
Figure 3. Crystal structure of [(Ph2PCH2CH2PPh2)CuCl]3 (16)
(hydrogens omitted for clarity).

Copper(I) Diphosphine Catalysts
Organometallics, Vol. 27, No. 13, 2008 3199
Table 6. Catalytic Amidation Results
runa
catalyst
yieldb (%)
runa
catalyst
yieldb (%)
1
13
CuI + 1
14
CuI + 2
15
CuI + 3
16
CuI + 4
17
CuI + 5
18
CuI + 6
19
CuI + 7
20
10
11
11
15
10
10
6
7
5
11
11
6
6
2
10
8
6
18c
19
CuI +9
22
CuI + 10
CuI + 10 + Et3N
23
6
9
2c
3
20c
21d
22
34
29
14
18
13
10
80
8
40
5
37
28
7
4c
5
6c
7
23c
24
CuI + 11
24
CuI + 12
8c
9
25c
26c
27
CuI + Me(H)N(CH2)2N(H)Me
10c
11
12c
13
14c
15
16c
17
CuI
CuCl + 10
CuCl
28e
29
30c
31c
32c
33f
34f
CuI + 25
CuI + 26
CuI + 27
CuI + 8
21
CuI + 10
34
31
CuI + 25
a Conditions: Reaction A: 2-iodotoluene (1.0 mmol), acetamide (1.5 mmol), K3PO4 (2.0 mmol), complexes 13-24 (0.05 mmol), 1 mL of DMF, 80
°C, 23 h. b Determined by GC, average of two runs. c Reaction A: As run 1 only CuI (0.05 mmol), ligands 1-12, 25-27 (0.1 mmol). d Reaction A: As
run 1 only CuI (0.05 mmol), ligand 10 (0.05 mmol), Et3N (0.5 mmol). e Reaction A: As run 1 only CuCl (0.05 mmol) and ligand 10 (0.1 mmol).
f Reaction B: 1,3-Dimethyl-5-iodobenzene (1.0 mmol), 2-pyrrolidinone (1.5 mmol), K3PO4 (2.0 mmol), ligands 10, 25 (0.1 mmol), CuI (0.05 mmol), 1
mL of toluene, 80 °C, 23 h.
Table 7. Crystallographic Details
13 · 2C2H4Cl2
14 · 2C2H4Cl2
colorless plate
0.40 × 0.40 × 0.10
C62H68Cl6Cu2P4
1276.82
16
17
23 · 3CH2Cl2
color, habit
size/mm
empirical formula
M
colorless block
0.10 × 0.08 × 0.04
C79H74Cl7Cu3P6
1647.97
colorless irregular block
0.4 × 0.3 × 0.3
C78H72Cl3Cu3P6
1492.15
colorless flat stalk
0.20 × 0.07 × 0.01
C30H32ClCuP2
553.49
colorless block
0.5 × 0.4 × 0.4
C61H68Cl6Cu4N2P4
1561.83
cryst syst
space group
triclinic
P1
triclinic
P1
monoclinic
P21/c
monoclinic
P21/n
triclinic
P1
j
j
j
a/Å
b/Å
c/Å
13.596(3)
14.178(3)
19.478(4)
90.24(3)
90.30(3)
90.96(3)
3754.2(13)
2
4.822
11.886(2)
12.866(2)
22.334(4)
81.31(4)
78.59(5)
63.63(4)
2991.6(9)
2
1.125
26.080(4)
15.074(3)
19.347(5)
90
109.63(3)
90
7164(3)
4
1.171
100
9.0865(18)
21.048(4)
14.503(3)
90
100.17(3)
90
2730.1(10)
4
1.032
14.551(3)
14.640(3)
16.927(3)
86.56(3)
79.15(3)
64.59(3)
3198.1(14)
2
1.7
R/deg
ꢀ/deg
γ/deg
V/Å3
Z
µ/mm-1
T (K)
100
173
100
173
reflns: total/indep/Rint
final R1 and wR2
largest peak, hole
29 546/12 807/0.0592
0.0530, 0.1503
1.235, -0.687
1.458
26 598/10 544/0.0962
0.0532, 0.1114
0.474, -0.510
1.42
80 379/16400/0.0789
0.0841, 0.2210
1.764, -0.683
1.384
30 759/6252/0.1357
0.0921, 0.1774
0.680, -0.500
1.347
34 227/14 518/0.0477
0.0432, 0.0957
0.549, -0.615
1.445
F
calc/g cm-3
a 12-membered-ring structure (Figure 2, Table 1). The structure
also contains three chloride ligands: one terminal (Cl3) coor-
dinated to Cu3, one that bridges Cu1 and Cu2 (Cl1), and one
that bridges all three copper atoms (Cl2), so that all three copper
atoms are tetrahedrally coordinated. The three dppm ligands,
the three copper atoms, and the µ3-bridging chloride make a
fused ring system containing three six-membered rings. The ring
containing Cu1, Cu3, P3, and P4 adopts a chair conformation,
as does that containing Cu1, Cu2, P1, and P2, which also has
the µ2-1,3-bridging chloride. The third ring, containing Cu2,
Cu3, P5, and P6 adopts an envelope conformation where five
of the atoms are essentially planar and the sixth atom (P5) lies
out of this plane. The Cu-Cl bond lengths vary from 2.31 Å
(Cu3-Cl3) to 2.47 Å (Cu3-Cl2).
ever, the different bridging ligand (dppe in 16, cf. dppm in 13)
and the extended conformation it adopts result in long Cu3 · · · Cu
distances and two µ2-bridging chlorides on the Cu1 · · · Cu2 edge
of the Cu3 triangle. Both Cu1 and Cu2 are tetrahedrally
coordinated, while Cu3, which is trigonal, carries one terminal
chloride ligand. The C2 backbones of the two dppe ligands
complexed to Cu3 both have anti conformations (P3-C39-
C40-P4 ) 172° and P5-C65-C66-P6 ) 167°), whereas the
dppe ligand bridging Cu1 and Cu2 adopts a distorted gauche
conformation (P1-C13-C14-P2 ) -113°). The Cu-Cl bond
length of the terminal chloride is much shorter (2.21 Å) than
the equivalent bond length in the crystal structure of 13 and is
likely a reflection of the lower coordination number of the
copper(I) in 16.
2.1.2. Crystal Structure of [(Ph2PCH2CH2PPh2)CuCl]3 (16).
Complex 16 has a similar coordination geometry to 13; each
pair of the three copper(I) atoms is bridged by a bis(diphe-
nylphosphino)ethane (dppe) ligand (Figure 3, Table 2). How-
2.1.3. Crystal Structure of [(Ar2PCH2PAr2)CuCl]2 · 2C2H4-
Cl2 {Ar ) 2-C6H4(Me)} (14). The crystal structure of the
j
dichloroethane solvate 14 has the space group P1 with one
molecule of complex and two molecules of dichloroethane in

3200 Organometallics, Vol. 27, No. 13, 2008
Daly et al.
suggesting the same active species is formed in either case.
Results are presented in Table 6.
With one notable exception, similarly low activity is seen in
all cases, in general preformed complexes 13-24 giving 5-14%
yields and in situ catalysts 2-18%. The similarity of these
values across a range of ligand structures leads to the conclusion
that catalysis is not being influenced by these particular ligands,6
corroborated by the similar result obtained in run 27, when only
CuI is used (8%). The exception to this trend is the N,N-
bis(diphenylphosphino)amine ligand 10 (run 20), which, when
reacted with CuI in situ, gives an improved yield of 34%. This
is double the figure of merit for other ligands, although still
some way short of the value obtained with the previously
reported diamine catalysts (run 26). Changing the metal salt to
CuCl (run 28) gives a yield of 40%. This is similar to the CuI
result, suggesting the same active species is formed in each case,
and is again significantly higher than the metal salt alone (run
29: 5%). Performance being independent of halide used is in
line with previous studies for copper diamine systems.5b The
reason for the higher activity of a catalyst system formed in
situ compared to using the preformed complex 22 is not clear,
although solubility issues cannot be ruled out. Increasing the
ortho aryl steric bulk of this diphosphine in 11 again yields a
catalyst in which ligand control is not exerted. Experiments in
which 1 equiv of 10 was used in conjunction with 1 equiv of
triethylamine rather than 2 equiv of 10 gave similar results (run
21).
Figure 7. N,N-Bis(diarylphosphino)amine ligands used.
the asymmetric unit. In this structure, the two diphosphine
ligands bridge a pair of copper(I) atoms, forming an eight-
membered ring (Figure 4, Table 3). Each copper carries one
terminal chloride ligand, and trigonal coordination results. The
Cu-Cl bond lengths are both around 2.23 Å. All four di-o-
tolylphosphino moieties are oriented such that one o-tolyl is
close to anti and one is close to gauche with respect to the
metal.
2.1.4. Crystal Structure of [(Ar2PCH2CH2PAr2)CuCl] {Ar
) 2-C6H4(Me)} (17). Complex 17 contains a trigonal copper(I)
atom coordinated by terminal chloride and chelating dppe
derivative ligands (Figure 5, Table 4). The five-membered
chelate ring adopts a δ twist conformation, with the torsion angle
P1-C1-C2-P2 ) 49°. Both o-tolyl groups on P2 are
disordered and have been modeled as lying over two orientations
about the P-C bond. One o-tolyl ring (C17) is disordered such
that the torsion angle Cu-P2-C17-C18 is either -14° (about
67%) or 174° (the remainder). The tolyl group at C24 shows a
slight disorder, which probably results from the disorder in the
neighboring tolyl group. The two tolyl groups on P1 have torsion
angles Cu1-P1-C3-C4 and Cu1-P1-C10-C11 ) -25° and
-67°. The Cu-Cl bond length is 2.17 Å, which is shorter than
the terminal chloride bond lengths in either 13 or 16.
This promising result for ligand 10 led us to screen several
other N,N-bis(diarylphosphino)amines (Figure 7). Similar results
are obtained in the amidation of an aryl iodide as the N-
substituent of the ligand is increased in bulk (runs 30-31);
however, increasing the bulk of the P-aryl ortho substituent has
a detrimental effect on catalysis, results now matching the
ligand-free system (run 32). Ligands 10 and 25 were also found
to be active in amidation reaction B (Table 6, runs 33-34).
2.1.5. Crystal Structure of [(Ar2PN(Me)PAr2)Cu2Cl2]2 ·
3CH2Cl2 {Ar ) 2-C6H4(Me)} (23). The dichloromethane solvate
j
23 crystallizes in the space group P1 with one molecule of
complex and approximately three molecules of dichloromethane
in the asymmetric unit. The complex consists of a distorted
octahedron of four copper(I) atoms and two chloride atoms, with
the four copper atoms in one plane and the chlorides in apical
sites (Figure 6, Table 5). Pairs of copper atoms on each edge
around the plane are bridged either by a further chloride or by
a diphosphine ligand. The two chlorine atoms that lie at the
apexes of the octahedron are not fully coordinated to all four
copper atoms; the Cu-Cl distances vary from 2.30 to 3.19 Å,
the latter being beyond a normal bond length. Both of these
chloride atoms have two “short” Cu-Cl distances (in the range
2.30-2.50 Å) and two longer distances. The Cu-Cu distances
vary from 2.75 to 2.87 Å, close to twice the van der Waals
radius of copper. The nitrogen that links the phosphorus atoms
in the bridging diphosphine ligand is planar (N2 sum of angles
around nitrogen 360.0°) or very close to planar (N1, 358.9°),
indicating delocalization of the lone pair.
2.2. Catalytic C-N Bond Formation. As a test reaction, the
catalytic coupling of 2-iodotoluene and acetamide under previ-
ously reported conditions was screened with complexes
13-24.5b In order to benchmark our systems against known
catalysts, runs in which a methodology reminiscent of these
previous reports, formation of a catalyst in situ with CuI and
ligands 1-12, were also attempted. Although the structural
chemistry of copper(I) iodide complexes is likely to be different
from copper(I) chloride complexes of the same ligands, in both
cases halide must be replaced to generate the Cu(I) monoam-
idinate species, which Buchwald and co-workers have demon-
strated to be the active catalysts in such reactions.5c Our results
(Vide infra) and those of others with N-donor ligands indicate
the same trend in results irrespective of the halide used,
3. Conclusions
Copper complexes of N,N-bis(diphenylphosphino)amine ligands
show moderate activity for amidation reactions, the first reported
copper phosphine systems to do so. This is in contrast to other
diphosphines which give only, within error, the same yields
achieved in the absence of any phosphine. It is noteworthy that
N,N-bis(diphenylphosphino)amines are also the most successful
diphosphines for nickel-catalyzed ethene polymerization, another
area where hard donor ligands predominate. We have previously
ascribed the distinctive performance of these ligands to a
combination of their small and rigid chelates and their unusual
electronic structure, the formal lone pair of the backbone
nitrogen being delocalized over the entire P-N-P backbone.
Buchwald and co-workers have demonstrated that chelating
ligands are crucial in their systems by preventing multiple
amidation and facilitating the formation of the active Cu(I)
monoamidinate species. It is plausible that the role of the
diphosphine ligand in this study is similarly to generate a Cu(I)
monoamidinate species under catalytic conditions. Given the
variations in structure observed for copper(I) chloride complexes
by subtle modification of the diphosphine used, the specific
ligand utilized is likely to have a crucial role in this regard.
4. Experimental Section
4.1. General Comments. All procedures were carried out under
an inert (N2) atmosphere using standard Schlenk line techniques
or in an inert atmosphere (Ar) glovebox. Chemicals were obtained

Copper(I) Diphosphine Catalysts
Organometallics, Vol. 27, No. 13, 2008 3201
from Sigma Aldrich, Fisher Scientific, Acros, Fluka, or Alfa Aesar
and used without further purification unless otherwise stated. All
solvents were purified using an Anhydrous Engineering Grubbs-
type solvent system.13 Ligands 1, 4, and 7 are commercially
available. Ligands 2, 3, 5, 6, 8-12, and 25-27 were synthesized
according to previously published methods.8
NMR spectra were recorded on JEOL GX270, ECP 300, Lambda
300, or GX400 spectrometers, 1H NMR chemical shifts are
referenced relative to the residual solvent resonances in the
dueterated solvent, and 31P{1H} NMR spectra are referenced
relative to high frequency of 85% H3PO4. Mass spectra were
recorded on a VG Analytical Quattro spectrometer. Microanalyses
were carried out by the Microanalytical Laboratory of the School
of Chemistry at the University of Bristol.
Synthesis of [(Ph2P(CH2)2PPh2)CuCl] (16). This was synthe-
sized in essentially the same way as complex 13 using ligand 4
(415 mg, 1.04 mmol) and copper(I) chloride (103 mg, 1.04 mmol)
to give 16 as a white solid. Yield: 495 mg, 96%. X-ray-quality
crystals of [(4)3Cu3Cl3] were formed by cooling a toluene solution
1
of 16. H NMR (CDCl3, 300 MHz): δ 2.43 (br, 4H, CH2), 7.09
(m, 8H, ArH); 7.21 (m, 12H, ArH). 31P{1H} NMR (CDCl3, 121
MHz): δ -9.8 (s). Anal. Calcd for C26H24CuClP2 (%): C 62.8, H
4.9. Found: C 65.4, H 5.1. MS (ESI): 959 [(4)2Cu2Cl]+, 859
[(4)2Cu]+.
Synthesis of [(2-C6H4(Me))2P(CH2)2P(2-C6H4(Me))2)CuCl] (17).
A solution of 5 (153 mg, 0.34 mmol) in toluene (5 mL) was added
to a suspension of copper(I) chloride (33 mg, 0.34 mmol) in toluene
(5 mL). The mixture was stirred overnight, and during this time a
white solid precipitated. The solid was isolated by filtration, washed
with hexane (2 × 5 mL), and dried in Vacuo to give 17 as a white
solid. Yield: 115 mg, 62%. X-ray-quality crystals of [(5)CuCl] were
formed by cooling a toluene solution of 17.
4.2. Synthesis of Copper(I) Diphosphine Chloride Complexes:
13-24. Synthesis of [(Ph2PCH2PPh2)CuCl] (13). A solution of 1
(373 mg, 0.97 mmol) in toluene (20 mL) was added to solid
copper(I) chloride (96 mg, 0.97 mmol), and the resulting mixture
was stirred for 20 h at room temperature. During this time a white
precipitate formed. The solvent was removed under reduced pressure
to give a pale pink solid. 13 was obtained by recrystallization from
a mixture of toluene and hexane as a white microcrystalline solid.
This was isolated by filtration, washed with 2 × 5 mL of hexane,
and dried in Vacuo. Yield: 469 mg, 75%. X-ray-quality crystals of
[(1)3Cu3Cl3] · 2C2H4Cl2 were formed from 1,2-dichloroethane/
hexane.
1H NMR (CDCl3, 400 MHz): δ 2.35 (t, 4H, CH2), 2.43 (s, 12H,
ArCH3), 7.11 (m, 8H, ArH), 7.26 (m, 8H, ArH). 31P{1H} NMR
(CDCl3, 121 MHz): δ -28.7 (s). Anal. Calcd for C30H32CuClP2
(%): C 65.1, H 5.8. Found: C 65.0 H 5.7. MS (ESI): 1071
[(5)2Cu2Cl]+.
Synthesis of [(2-C6H4(i-Pr))2P(CH2)2P(2-C6H4(i-Pr))2)CuCl] (18).
This was synthesized in essentially the same way as complex 13
using ligand 6 (201 mg, 0.35 mmol) and copper(I) chloride (35
mg, 0.35 mmol) to give 18 as a white microcrystalline solid. Yield:
142 mg, 61%.
1H NMR (CDCl3, 400 MHz): δ 3.21 (s, 2H, CH2), 6.95 (m, 8H,
ArH), 7.11 (m, 12H, ArH). 31P{1H} NMR (CDCl3, 121 MHz): δ
-14.4 (s). Anal. Calcd for C25H22CuClP2 (%): C 62.1, H 4.6. Found:
C 58.6, H, 4.5. This composition is consistent with the empirical
formula C25H22CuClP2 · 0.5C2Cl2H4 (C 58.6, H 4.5), 0.5 mol of
solvent present for every mol of [(1)CuCl] in the crystal structure.
MS (ESI): 1413 [(1)3Cu3Cl2]+.
Synthesis of [(2-C6H4(Me))2PCH2P(2-C6H4(Me))2)CuCl] (14).
This was synthesized in essentially the same way as complex 13
using ligand 2 (200 mg, 0.45 mmol) and copper(I) chloride (45
mg, 0.45 mmol) to give 14 as a white microcrystalline solid. Yield:
230 mg, 76%. X-ray-quality crystals of [(2)2Cu2Cl2] · 2C2H4Cl2 were
formed from 1,2-dichloroethane/hexane.
1H NMR (CDCl3, 300 MHz): δ 0.88 (d, 12H, CH(CH3)2), 0.96
(d, 12H, CH(CH3)2), 2.33 (t, 4H, CH2), 3.71 (m, 4H, CH(CH3)2),
7.09 (m, 4H, ArH), 7.28 (m, 12H, ArH). 31P{1H} NMR (CDCl3,
121 MHz): δ -31.7 (s). Anal. Calcd for C38H48CuClP2 (%): C 68.6,
H 7.3. Found: C 68.4, H 7.2. MS (ESI): 1353 Na+[(6)2Cu2Cl2],
1295 [(6)2Cu2Cl]+.
Synthesis of [(Ph2P(CH2)3PPh2)2Cu]Cl (19). A solution of 7 (223
mg, 0.54 mmol) in toluene (20 mL) was added to solid copper(I)
chloride (53 mg, 0.54 mmol), and the resulting mixture was stirred
for 4 h. The solvent was removed under reduced pressure, and the
residue was triturated with diethyl ether (2 × 10 mL) to give 19 as
a white solid. 19 was recrystallized from hot acetonitrile. Yield:
235 mg, 94%.
1H NMR (CDCl3, 270 MHz): δ 1.66 (v br, 8H, ArCH3), 2.03
(br, 4H, ArCH3), 3.62 (br, 2H, CH2), 6.64-7.43 (br m, 16H, ArH).
31P{1H} NMR (CDCl3, 121 MHz): δ -18.4 (s). Anal. Calcd for
C29H30CuClP2 (%): C 64.6, H 5.6. Found: C 59.9, H 5.4. This
composition is consistent with the empirical formula
C29H30CuClP2 · 0.75C2Cl2H4 (C 59.7, H 5.4), 1,2-dichloroethane
present in the crystal structure. MS (ESI): 1043 [(2)2Cu2Cl]+.
Synthesis of [(2-C6H4(i-Pr))2PCH2P(2-C6H4(i-Pr))2)Cu2Cl2] (15).
A solution of 3 (106 mg, 0.19 mmol) in toluene (15 mL) was added
to solid copper(I) chloride (19 mg, 0.19 mmol), and the resulting
mixture was stirred for 20 h at room temperature. The reaction was
monitored by 31P{1H} NMR, which showed a large amount of
unreacted ligand (δ ) -49.5 (s), -22.0 (br, s)). The mixture was
heated at reflux for several days without any change in the 31P NMR
spectrum. The solvent was removed under reduced pressure, and the
resulting solid was washed with hexane to remove the excess ligand.
The white solid 15 was dried in Vacuo. Yield: 45 mg, 63%.
1H NMR (CDCl3, 400 MHz): δ 0.66 (d, 12H, CH(CH3)2), 0.85
(d, 12H, CH(CH3)2), 3.16 (t, 2H, CH2), 3.70 (m, 4H, CH(CH3)2)
6.91 (t, 4H, ArH), 7.10-7.40 (m, 10H, ArH). 31P{1H} NMR
(CDCl3, 121 MHz): δ -17.7 (s, br). Anal. Calcd for C37H46Cu2Cl2P2
(%): C 59.2, H 6.2. Found: C 60.6, H 6.4. MS (ESI): 1465
[(3)2Cu4Cl3]+.
1H NMR (CDCl3, 270 MHz): δ 1.83 (br, 2H, CH2), 2.38 (br,
4H, PCH2), 7.10-7.79 (m, 20H, ArH). 31P{1H} NMR (CDCl3, 121
MHz): δ -5.0 (br), –12.0 (br). Anal. Calcd for C54H52CuClP4 (%):
C 70.2, H 5.7. Found: C 70.5, H 5.6. MS (ESI): 887 [(7)2Cu]+.
Synthesis of [(2-C6H4(Me))2P(CH2)3P(2-C6H4(Me))2)CuCl] (20).
This was synthesized in essentially the same way as complex
17 using ligand 8 (100 mg, 0.213 mmol) and copper(I) chloride
(21 mg, 0.213 mmol) to give 20 as a white solid. Yield: 90 mg,
74%.
1H NMR (CD2Cl2, 300 MHz): δ 1.92 (m, 2H, CH2), 2.37
(apparent s, 16H, ArCH3, and PCH2 obscured), 7.11-7.31 (m, 16H,
ArH). 31P{1H} NMR (CDCl3, 121 MHz): δ -27.9 (s). Anal. Calcd
for C31H34CuClP2 (%): C 65.6, H 6.0. Found: C 65.6, H 6.2. MS
(ESI): 1099 [(8)2Cu2Cl]+, 631 [(8)Cu2Cl]+, 531 [(8)Cu]+.
Synthesis of [(2-C6H4(i-Pr))2P(CH2)3P(2-C6H4(i-Pr))2)CuCl] (21).
This was synthesized in essentially the same way as complex
17 using ligand 9 (226 mg, 0.389 mmol) and copper(I) chloride
(39 mg, 0.389 mmol) to give 21 as a white solid. Yield: 238
mg, 90%.
1H NMR (CDCl3, 270 MHz): δ 0.84 (d, 12H, CH(CH3)2), 1.01
(d, 12H, CH(CH3)2), 2.03 (m, 2H, CH2), 2.46 (br s, 4H, PCH2),
3.68 (m, 4H, CH(CH3)2), 7.11-7.46 (m, 16H, ArH). 31P{1H} NMR
(CDCl3, 121 MHz): δ -28.1 (s). Anal. Calcd for C39H50CuClP2
(11) Camus, A.; Marsich, N.; Nardin, G.; Randaccio, L. Inorg. Chim.
Acta 1977, 23, 131.
(12) (a) Mehrotra, P. K.; Hoffmann, R. Inorg. Chem. 1978, 17, 2187.
(b) Hermann, H. L.; Boche, G.; Schwerdtfeger, P. Chem.-Eur. J. 2001, 7,
5333.
(13) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics. 1996, 15, 1518.

3202 Organometallics, Vol. 27, No. 13, 2008
Daly et al.
(%): C 68.9, H 7.4. Found: C 71.5, H 7.7. This composition is
consistent with the empirical formula C39H50CuClP2 · C7H8 (C 71.5,
H 7.7) with toluene observed in the 1H NMR spectrum. MS (ESI):
1323 [(9)2Cu2Cl]+.
in calculated positions, with isotropic displacement parameters equal
to 1.5 times (methyl hydrogen atoms) or 1.2 times (all other
hydrogen atoms) that of their attached atom. Complex neutral-atom
scattering factors were used.18
Synthesis of [(Ph2PN(Me)PPh2)CuCl] (22). This was synthesized
in essentially the same way as complex 17 using ligand 10 (100
mg, 0.25 mmol) and copper(I) chloride (25 mg, 0.25 mmol) to give
22 as a white solid. Yield: 60 mg, 48%.
In 13, one phenyl ring (C64-C69) is disordered and has been
modeled as lying over two orientations. The displacement ellipsoids
for one of the other phenyl rings (C14-C19) indicate that it may
also be disordered, but no satisfactory model with two or more
ring positions could be obtained. The chlorine atoms in one of the
dichloroethane solvent molecules are also disordered and have been
modeled as lying over two positions.
1H NMR (CDCl3, 270 MHz): δ 1.98 (br, 3H, NCH3), 6.99-7.30
(m, 20H, ArH). 31P{1H} NMR (CDCl3, 121 MHz): δ 58.5 (s); Anal.
Calcd for C25H23CuClNP2 (%): C 60.3, H 4.7, N 2.8. Found: C
58.4, H 4.8, N 2.8. MS (ESI): 1458 [(10)3Cu3Cl]2+
.
The crystal structure of 23 contained residual electron density
that could not be satisfactorily identified and was modeled using
the SQUEEZE algorithm incorporated into the PLATON suite. This
found an extra 78 electrons in a void of 278 Å3, which corresponds
to approximately two further molecules of dichloromethane in the
asymmetric unit in addition to one ordered solvent molecule. Except
as discussed above, refinement proceeded smoothly to give the
shown structures.
4.4. Catalytic Screening. Coupling of 2-Iodotoluene with Ac-
etamide. A Schlenk tube was charged with acetamide (90 mg, 1.52
mmol) and K3PO4 (430 mg, 2.03 mmol). The tube was evacuated
and backfilled with nitrogen several times, and then either copper(I)
iodide (9.6 mg, 0.05 mmol) or copper(I) chloride (5.0 mg, 0.05
mmol) and the appropriate ligand (0.10 mmol) or the appropriate
preformed copper(I) chloride complex (0.05 mmol) was added
followed by 2-iodotoluene (128 µL, 1.01 mmol) and dimethylfor-
mamide (1.0 mL). The tube was heated to 80 °C, sealed, and stirred
for 23 h. The resulting suspension was allowed to cool to room
temperature and filtered through a 0.5 × 1 cm plug of silica gel,
eluting with 10 mL of ethyl acetate. Mesitylene (139 µL, 1.00
mmol) was added as a standard, and the solution was analyzed by
GC.
Synthesis of [(2-C6H4(Me))2PN(Me)P(2-C6H4(Me))2)CuCl] (23).
Reaction in toluene: This was synthesized in essentially the same
way as 17 using 11 (54 mg, 0.12 mmol) and copper(I) chloride
(12 mg, 0.12 mmol) to give 23 as a pale yellow solid. Yield: 39
mg, 59%. X-ray-quality crystals of [(11)2Cu4Cl4] · 3CH2Cl2 were
formed from dichloromethane.
1H NMR (CD2Cl2, 300 MHz): δ 2.09 (br, 6H, ArCH3), 2.30 (br,
6H, ArCH3), 2.82 (br, 3H, NCH3), 6.83-7.42 (br m, 16H, ArH).
31P{1H} NMR (CDCl3, 121 MHz): δ 55.7 (s). Anal. Calcd for
C29H31CuCl2NP2 (%): C 59.0, H 5.3, N 2.4. Found: C 59.8, H 5.7,
N 2.7. MS (ESI): 1271 [(11)2Cu4Cl3]+, 973 [(11)2Cu]+.
Synthesis of [(2-C6H4(i-Pr))2PN(Me)P(2-C6H4(i-Pr))2)CuCl] (24).
A solution of 12 (70 mg, 0.123 mmol) in toluene (5 mL) was added
to a solution of copper(I) chloride (12 mg, 0.123 mmol) in toluene,
and the resulting mixture was stirred overnight. The solvent was
removed under reduced pressure, and the residue was triturated with
hexane (2 × 5 mL) to give 24 as a white solid. Yield: 64 mg,
74%.
1H NMR (CDCl3, 300 MHz): δ 0.60-1.34 (br m, 24H,
CH(CH3)2), 2.43-2.93 (br m, 3H, NCH3), 3.24-3.82 (br m, 4H,
CH(CH3)2), 6.74-7.58 (br m, 16H, ArH). 31P{1H} NMR (CDCl3,
121 MHz): δ 61.9 (s). Anal. Calcd for C37H47CuClNP2 (%): C 66.7,
H 7.1, N 2.1. Found: C 64.6, H 7.5, N 2.0. MS (ESI): 1460
Coupling of 1,3-Dimethyl-5-iodobenzene with 2-Pyrrolidinone.
A Schlenk tube was charged with copper(I) iodide (9.6 mg, 0.05
mmol) and K3PO4 (430 mg, 2.03 mmol). The tube was evacuated
and backfilled with nitrogen several times, and the appropriate
ligand (0.10 mmol) was added followed by 2-pyrrolidinone (116
µL, 1.52 mmol), 1,3-dimethyl-5-iodobenzene (146 µL, 1.01 mmol),
and toluene (1.0 mL). The tube was heated to 80 °C, sealed, and
stirred for 23 h. The resulting suspension was allowed to cool to
room temperature and filtered through a 0.5 × 1 cm plug of silica
gel, eluting with 10 mL of ethyl acetate. Mesitylene (139 µL, 1.00
mmol) was added as a standard, and the solution was analyzed by
GC.
[(12)2Cu4Cl2]2+
.
4.3. X-ray Crystallography. X-ray diffraction experiments on
16 and 17 were carried out at 100 K on a Bruker SMART APEX
diffractometer, and experiments on 14 and 23 were carried out at
173 K on a Bruker SMART diffractometer, both using Mo KR
X-radiation (λ ) 0.71073 Å). A Bruker PROTEUM diffractometer
using Cu KR radiation (λ ) 1.54157 Å) was used for 13. All
experiments were performed using a single crystal coated in paraffin
oil mounted on a glass fiber.14 All three diffractometers used a
CCD area detector, and intensities were integrated15 from several
series of exposures, each exposure covering 0.3° in ω. Absorption
corrections were based on multiple and symmetry-equivalent
measurements done with SADABS V2.10.16 Structures were refined
using SHELXTL17 against all Fo2 data with hydrogen atoms riding
Acknowledgment. We would like to thank the EPSRC
and University of Bristol for funding.
Supporting Information Available: Crystallograpic data (cif
files). This material is available free of charge via the Internet at
http://pubs.acs.org.
(14) SMART diffractometer control software. SMART collections: version
5.054; Bruker-AXS Inc.: Madison, WI, 1997-1998; SMART APEX col-
lections, version 5.628; Bruker-AXS Inc.: Madison, WI, 2005; PROTEUM
collections, version 5.625; Bruker-AXS Inc.: Madison, WI, 1997-2001.
(15) SAINT V6.02-7.06A integration software; Siemens Analytical X-ray
Instruments Inc.: Madison, WI.
OM800139V
(16) Sheldrick, G. M. SADABS V2.10; University of Go¨ttingen, 2003.
(17) SHELXTL program system version 5.1; Bruker Analytical X-ray
Instruments Inc.: Madison, WI, 1998.
(18) International Tables for Crystallography; Kluwer: Dordrecht, 1992;
Vol. C.