The activation of tertiary carboxamides in metal complexes: An experimental and theoretical study on the methanolysis of acylated bispicolylamine copper(II) complexes

Article abstract of DOI:10.1021/ic0496774

It is a well-established concept that the C-N bond cleavage of carboxamide functions is facilitated by the coordination of a metal ion to the carbonyl oxygen atom. In contrast, the alternative C-N bond activation by coordination of a neutral tertiary carboxamide nitrogen atom has not been studied. We present the first results on the effect of nitrogen pyramidalization in N-coordinated metal complexes on the methanolysis of tertiary carboxamide groups. An analysis of the reactions products obtained from the methanol cleavage of [(N-Acyl-bpa)Cu]²⁺ (bpa = N,N-bispicolylamine) complexes is presented together with experimental and high-level theoretically calculated structures. The strong effect of different anions on the amide pyramidalization and subsequent C-N-bond cleavage is evaluated. We show that dichloro complexes [(N-Acyl-bpa)CuCl₂] have much less activated amide groups than the corresponding triflate species. They should therefore be less reactive. However, [(N-Acyl-bpa)CuCl₂] complexes dissociate in solution to give cationic monochloro complexes [(N-Acyl-bpa)Cu(S)Cl]⁺ (S = solvent molecule). Theoretical calculations show that the amide pyramidalization in the monochloro complexes is equal to that in the corresponding CF₃SO ₃^- salts. Consequently, chloro and triflato complexes are cleaved with similar rates and efficiencies. Parallels to and differences in the reactivity of purely organic distorted amides are discussed.

Full text of DOI:10.1021/ic0496774

Inorg. Chem. 2004, 43, 4663−4673
The Activation of Tertiary Carboxamides in Metal Complexes:
An Experimental and Theoretical Study on the Methanolysis of Acylated
Bispicolylamine Copper(II) Complexes
,§
,†
Nicole Niklas,† Frank W. Heinemann,† Frank Hampel,‡ Tim Clark,* and Ralf Alsfasser*
Institute of Inorganic Chemistry, UniVersity of Erlangen-Nu¨rnberg, Egerlandstr. 1,
D-91058 Erlangen, Germany, Computer-Chemie-Centrum, Na¨gelsbachstrasse 25,
D-91052 Erlangen, Germany, and Institute of Organic Chemistry, UniVersity of
Erlangen-Nu¨rnberg, Henkestrasse 42, D-91054 Erlangen, Germany
Received March 12, 2004
It is a well-established concept that the C−N bond cleavage of carboxamide functions is facilitated by the coordination
of a metal ion to the carbonyl oxygen atom. In contrast, the alternative C−N bond activation by coordination of a
neutral tertiary carboxamide nitrogen atom has not been studied. We present the first results on the effect of
nitrogen pyramidalization in N-coordinated metal complexes on the methanolysis of tertiary carboxamide groups.
An analysis of the reactions products obtained from the methanol cleavage of [(N-Acyl-bpa)Cu]2+ (bpa ) N,N-
bispicolylamine) complexes is presented together with experimental and high-level theoretically calculated structures.
The strong effect of different anions on the amide pyramidalization and subsequent C−N-bond cleavage is evaluated.
We show that dichloro complexes [(N-Acyl-bpa)CuCl2] have much less activated amide groups than the corresponding
triflate species. They should therefore be less reactive. However, [(N-Acyl-bpa)CuCl2] complexes dissociate in
solution to give cationic monochloro complexes [(N-Acyl-bpa)Cu(S)Cl]+ (S ) solvent molecule). Theoretical calculations
show that the amide pyramidalization in the monochloro complexes is equal to that in the corresponding CF SO -
3
3
salts. Consequently, chloro and triflato complexes are cleaved with similar rates and efficiencies. Parallels to and
differences in the reactivity of purely organic distorted amides are discussed.
Introduction
hybridization at the N-atom and concomitant weakening of
the amide-resonance stabilization, which should lead to a
weakening of the C-N bond. This mechanism has been
proposed3 but until now there has been no way to study it,
even though a comparison with the well-known oxygen
activation would be of fundamental interest for a better
understanding of carboxamide coordination chemistry. The
lack of information has a simple reason. For a long time
N-coordinated tertiary carboxamides could only be found in
the literature as hypothetical intermediates needed to ratio-
nalize the metal-mediated cleavage of strained amides such
as penicillins,3 or enhanced cis-trans-isomerization rates
about the C-N bond in the presence of metal ions such as
Ag+.4 These suggestions were controversial since other
The metal-induced cleavage of carboxamides is a well-
established reaction. Its importance in hydrolytic enzymes1
inspired sophisticated biomimetic applications such as the
site-selective cleavage of peptides by synthetic palladium
complexes.2 In all well understood cases, the amide bond is
activated by coordination of the amide carbonyl oxygen atom
to the metal center. However, the coordination of an
electrically neutral nitrogen atom in metal complexes of
tertiary carboxamides is an alternative that has gained little
attention. This mode of coordination should result in sp3
* To whom correspondence should be addressed. E-mail: alsfasser@
chemie.uni-erlangen.de (R.A.); clark@chemie.uni-erlangen.de (T.C.). Fax:
++49-9131-8527387 (R.A.); ++49-9131-8526565 (T.C.).
† Institute of Inorganic Chemistry, University of Erlangen-Nu¨rnberg.
‡ Institute of Organic Chemistry, University of Erlangen-Nu¨rnberg.
§ Computer-Chemie-Centrum.
(1) Proteolytic Enzymes, 2nd ed.; Beynon, R., Bond, J. S., Eds.; Oxford
University Press: New York, 2001.
(2) Milovic, N. M.; Kostic; N. M. J. Am. Chem. Soc. 2003, 125, 781 and
references therein.
(3) (a) Page, M. I. Acc. Chem. Res. 1984, 17, 144. (b) Constable, E. C.
Metals and Ligand ReactiVity; VCH: Weinheim, 1996; p 56.
(4) (a) Temessi, P. A.; Quadrifoglio, F. J. Chem. Soc., Chem. Commun.
1968, 844. (b) Armbruster, A. M.; Pullman, A. FEBS Lett. 1974, 49,
18. (c) Waghorne, W. E.; Ward, A. J. I.; Clune, T. G.; Cox, B. G. J.
Chem. Soc., Faraday Trans. 1 1980, 76, 1131.
10.1021/ic0496774 CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/01/2004
Inorganic Chemistry, Vol. 43, No. 15, 2004 4663

Niklas et al.
Scheme 1. Structures of Ligands 1a-d
Scheme 2. Structures of Copper(II) Complexes 2a-d and 3a-d
authors categorically ruled out the possible coordination of
a nondeprotonated amide nitrogen atom because of the more
favorable binding mode to the more basic carbonyl oxygen.5
More than 30 years ago, Houghton and Puttner6 studied the
methanolysis of N,N-bispicolylamide derivatives in the
presence of CuCl2. The idea was to use the thermodynami-
cally stable bispicolylamine (bpa) copper(II) complexes as
leaving groups in acylation reactions.7 However, neither the
nature of the actual reactive species nor the structures of the
products were reported. Only in 1996 Lectka et al. were able
to crystallize the benzoyl substituted derivative [(Benz-bpa)-
CuCl2], which provided the first example of an N-bound
tertiary carboxamide in a Werner-type complex.8 Over the
last three years, we have synthesized and characterized a
larger number of these unusual compounds and shown that
they contain pyramidal amide groups that are significantly
to strongly activated.9 This allows us to study the detailed
mechanism of nucleophilic amide cleavage reactions for the
first time. In particular, it was not possible to explain why
CuCl2 and Cu(CF3SO3)2 are similarly well suited to promote
the C-N bond methanolysis in acylated bpa derivatives
based solely on our structural data. Complexes of the latter
have much more strongly activated amide groups. We have
therefore combined an experimental evaluation of the
structures of these complexes and their products with ki-
netic data and the first theoretical calculations designed
to determine the relevant species during the CuCl2 and
Cu(CF3SO3)2 mediated cleavages of acylated bpa derivatives.
sulting compounds. The analytically pure triflate salts 2a-d
precipitated from dichloromethane-ether solvent mix-
tures. C,H,N-elemental analysis data indicate that samples
of [(Benz-bpa)Cu(CF3SO3)2] (2a) and [(Boc-Ala-bpa)-
Cu(CF3SO3)2] (2c) do not always contain a coordinated water
molecule.9 In analogy to a Cu(NO3)2 complex of bpa,11 as
well as a Cu(CF3SO3)2 complex of a diethylenetriamine
derivative,12 we assume square-pyramidal structures with two
coordinated triflate anions, one in an axial position and one
in an equatorial position. This is analogous to the square-
pyramidal amino acid derivatives 2b and 2d which have an
equatorial aqua and an axial triflato ligand. A crystal structure
was obtained for the glycine derivative 2b. The similarity
of 2d is indicated by spectroscopic and analytical data.
The CuCl2 complexes 3a-d were synthesized and stored
in the dark since prolonged exposure to light resulted in
decomposition, presumably due to the formation of chloro-
cuprate salts as was indicated by elemental analysis data of
the yellow-orange products. It is shown below that chloro-
cuprate(II) ions are easily formed in the investigated system.
We have also observed a similar behavior in related com-
plexes.13 All compounds precipitate from the reaction mix-
tures upon standing. We have crystallized and solved the
structure of 3c. The structure of 3a was determined by Lectka
et al.8 Crystals were also obtained for the leucine derivative
3d, but the X-ray diffraction data were not of publishable
quality yet nevertheless confirmed a structure similar to 3a
and 3c.
Results and Discussion
All the copper complexes readily undergo methanolysis
of their C-N-bonds to yield the methyl ester of the acyl
substituent and complexes of the [(bpa)Cu]2+ fragment. We
have studied four selected reactions on a preparative scale
quantitatively. They are summarized in Scheme 3.
The ligand Boc-Ala-bpa (1c) was reacted with Cu(CF3SO3)2
in pure methanol to give the bpa complex [(bpa)Cu(H2O)]-
[CF3SO3]2 (4) and Boc-Ala-OMe. Although we were not able
to crystallize 4, spectroscopic and analytical data indicate
that its structure is similar to that of the Jahn-Teller distorted
octahedral perchlorate derivative [(bpa)Cu(CH3OH)][ClO4]2.9
The two compounds have similar ESR parameters (120 K;
Synthesis and Reactivity Studies. Scheme 1 shows the
ligands Benz-bpa (1a),8,9 Boc-Gly-bpa (1b), Boc-Ala-bpa
(1c),9 and Boc-Leu-bpa (1d) which were synthesized from
bispicolylamine and the appropriate carboxylic acid accord-
ing to the DCC/HOBt coupling method (DCC, dicyclohexyl-
cabodiimide; HOBt, N-hydroxybenzotriazole).10
Copper complexes of 1a-d are readily obtained by
reaction of the appropriate ligand with Cu(CF3SO3)2 or
CuCl2‚2H2O in acetonitrile. Scheme 2 illustrates the re-
(5) (a) Freeman, H. C. Inorg. Biochem. 1973, 1, 121. (b) Sigel, H.; Martin,
R. B. Chem. ReV. 1982, 82, 385. Martin, R. B. Met. Ions Biol. Syst.
2001, 38, 1.
(6) Houghton, R. P.; Puttner, R. R. J. Chem. Soc., Chem. Commun. 1970,
1270.
(7) See also: Hay, R. W. In ComprehensiVe Coordination Chemistry;
Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.; Pergamon
Press: Oxford, 1987; Vol. 6, p 411.
(10) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis, 2nd
ed.; Springer-Verlag: Berlin, 1994.
(11) Palaniandavar, M.; Mahadevan, S.; Ko¨ckerling, M.; Henkel, G. J.
Chem. Soc., Dalton Trans. 2000, 1151.
(8) Cox, C.; Feraris, D.; Murthy, N. N.; Lectka, T. J. Am. Chem. Soc.
1996, 118, 5332.
(9) (a) Niklas, N.; Hampel, F.; Liehr, G.; Zahl, A.; Alsfasser, R. Chem.
Eur. J. 2001, 7, 5135. (b) Niklas, N.; Heinemann, F. W.; Hampel, F.;
Alsfasser, R. Angew. Chem., Int. Ed. 2002, 41, 3386.
(12) Scott, M. J.; Lee, S. C.; Holm, R. H. Inorg. Chem. 1994, 33, 4651.
(13) (a) Niklas, N.; Wolf, S.; Liehr, G.; Anson, C. E.; Powell, A. K.;
Alsfasser, R. Inorg. Chim. Acta 2001, 314, 126. (b) Niklas, N.; Hampel,
F.; Walter, O.; Liehr, G.; Alsfasser, R. Eur. J. Inorg. Chem. 2002,
1839.
4664 Inorganic Chemistry, Vol. 43, No. 15, 2004

Tertiary Carboxamides in Metal Complexes
Figure 1. (a) UV-vis spectra obtained during the methanolysis of 2a (10-2 M in methanol); (b) UV-vis spectra obtained during the methanolysis of 3a
(10-3 M in CH3OH/CH2Cl2 4:1).
Scheme 3. Preparative Studies on Amide Methanolysis Reactions
4, g| ) 2.26, A| ) 187 × 10-4 cm-1, g ) 2.07; ClO4- salt,
g| ) 2.26, A| ) 186 × 10-4 cm-1, g ) 2.07) and d-d-
absorption bands (4, λmax ) 661 nm, ꢀ ) 88 M-1 cm-1;
ClO4- salt, λmax ) 660 nm, ꢀ ) 89 M-1 cm-1) in methanol.
More interesting and revealing are the products formed
upon cleavage of the chloro complexes. Mixing the ligands
1a, 1b, or 1d with CuCl2‚2H2O in methanol resulted in the
immediate precipitation of the corresponding complexes 3a,
3b, and 3d. In the case of 1d, it was possible to dissolve the
complex by addition of methanol. The reaction of 1d in pure
methanol provided the best conditions for a quantitative
preparation of [(bpa)CuCl2] (5). Crystals of 5 were obtained,
and the structure is described below. The alanine derivative
1c also gave 5 from pure methanol solutions. In the case of
1a, 1b, and 1d, the complexes dissolved upon addition of
an equal volume of CH2Cl2 to methanol suspensions and
were subsequently cleaved. Prolonged standing of the
solutions always produced the dark blue complex [(bpa)-
CuCl2] (5). However, upon ether diffusion into the reaction
mixtures we were able to isolate a green byproduct together
with small amounts of the starting material and 5 after a
few hours. An X-ray structural analysis of this material
revealed the composition {[(bpa)CuCl2][(bpa)Cu(H2O)Cl]-
[Cu(bpa)Cl][CuCl4]‚CH3OH} (6). This was confirmed by
C,H,N-elemental analysis data and density measurements on
single crystals. The result is important because it provides
evidence for the formation of free copper(II)-chloro species
and monocationic LCuCl+ complexes in the course of the
reactions.
followed by UV-vis spectroscopy. Starting materials in these
studies were the analytically pure isolated complexes.
Addition of 180 µL of aqueous phosphate buffer (pH 7; 0.1
M) to 100 mL of the reaction mixtures assured comparable
water contents and the scavenging of potentially catalytic
trace amounts acid or base. Under these conditions, all results
were perfectly reproducible. The triflate salts were studied
in methanol solutions (10-2 M Cu2+) whereas 4:1 methanol/
dichloromethane mixtures were used in the case of the chloro
complexes (10-3 M Cu2+) because of their low solubility in
CH3OH. Figure 1 shows two typical experiments. Although
the data of 2a seem to show two perfect isosbestic points,
we were not able to fit the kinetic traces to a rate law.
Preliminary ESR data indicate that this is most probably due
to partial dissociation and other dynamic processes of the
complex. However, it is evident that the chloro complexes
and the triflate salts are both cleaved within ca. 8 h. This is
surprising since the structural data discussed below clearly
indicate that the dichloro species should have much less
activated amide groups than their triflate analogues. On the
basis of the structure of 6, we reasoned that monochloro
complexes could be the actual reactive species. We have
therefore performed a theoretical study to confirm that the
amide group in [(Ac-bpa)CuCl]+ is significantly more
distorted than in the dichloro complex.
X-ray Structural Analyses
[(Boc-Gly-bpa)Cu(H2O)(CF3SO3)](CF3SO3) (2b). The
structure of the complex 2b is shown in Figure 2. It consists
of the distorted square-pyramidal cation [(Boc-Gly-bpa)Cu-
(H2O)(CF3SO3)]+ and a loosely bound triflate ion (d(Cu-
In order to compare the different reactivities of the copper
complexes, the methanolysis of 2a-d and 3a-d was
Inorganic Chemistry, Vol. 43, No. 15, 2004 4665

Niklas et al.
Figure 3. ORTEP plots (30% ellipsoids) of the structures of (S,RP)-3c
(left) and (S,SP)-3c (right). Selected bond lengths and angles are listed in
Table 1.
Figure 2. ORTEP plot (30% ellipsoids) of the RP enantiomer of 2b.
Selected bond lengths and angles are listed in Table 1.
complexes, and only a few examples are known where
d(Cu-O) is between 2.3 and 2.4 Å.16 Shorter Cu-O(triflate)
distances than in 2b have only been reported for trigonal-
bipyramidal17 and tetrahedral18 complexes, or in compounds
were a triflato ligand is located in an equatorial position of
a square-pyramid. The latter case has been observed by Holm
et al. in a diethylenetriamine complex.12 Interestingly, the
authors find an axial Cu-triflate bond of 2.21 Å, similar to
that in 2b. Thus, the poor equatorial electron donation from
the N-bound amide function in 2b, or from the triflate ion
in Holm’s complex, is compensated by an unusually strong
axial copper to triflate bond.
Table 1. Selected Bond Lengths (Å) and Angles (deg) of the Tertiary
Amide Complexes 2b and 3c,a and the Calculated Structures
[(Ac-bpa)CuCl2] and [(Ac-bpa)CuCl(CH3OH)]+
3c‚0.5H2O‚CH2Cl2
[(Ac-bpa)- [(Ac-bpa)Cu-
CuCl2]
2b‚CH2Cl2 (S,RP)
(S,SP)
Cl(CH3OH)]+
Cu-N1
Cu-N2
Cu-N3
Cu-O4
2.164(3) 2.460(7) 2.437(6)
1.952(3) 1.991(7) 1.960(7)
1.948(3) 1.982(7) 1.968(7)
2.027(2)
2.627
2.058
2.074
2.222
2.018
2.045
Cu-Cl1
Cu-Cl2
Cu-O11
2.227(2) 2.227(2)
2.269(2) 2.269(2)
2.209(3)
2.286
2.262
2.363
2.239
Cu-O21
2.623(3)
[(Boc-Ala-bpa)CuCl2] (3c). Compound 3c crystallizes in
the noncentrosymmetric space group P21 with 4 independent
molecules per asymmetric unit. Their relevant geometric
parameters are very similar with one exception. The chiral
amino acid moiety in combination with the chirality plane
defined by the amide group gives rise to the formation of
diastereomers. Three molecules are in the (S,SP) conformation
whereas one is in the (S,RP) form. Both isomers are shown
in Figure 3, and bond parameters are given for both forms.
A comparison of the structure of 3c with those of the closely
related complexes 3a, which were reported by Lectka et al.,8
and the preliminary data obtained for 3d confirm that all
dichloro complexes are very similar. Their characteristic
features are two almost equal Cu-Cl distances, as well as
weak copper-amide bonds ca. 0.3 Å longer than in 2b and
ca. 0.5 Å longer than the copper to secondary amine bond
in common bpa complexes. The following discussion there-
fore applies to all known derivatives. However, it is
interesting to note that the leucine derivative 3d crystallizes
exclusively in the (S,SP) conformation.
N1-Cu-N2
N1-Cu-N3
N1-Cu-O4
N1-Cu-Cl1
N1-Cu-Cl2
N1-Cu-O11
N2-Cu-N3
N2-Cu-O4
N2-Cu-Cl1
N2-Cu-Cl2
N2-Cu-O11
N3-Cu-O4
N3-Cu-Cl1
N3-Cu-Cl2
N3-Cu-O11
O4-Cu-O11
82.8(2)
83.5(2)
165.4(2)
78.5(3) 78.0(2)
78.5(2) 78.9(3)
74.98
75.87
81.12
81.13
119.9(2) 120.1(2)
103.7(2) 101.9(2)
91.57
116.18
91.93
169.66
101.8(2)
165.2(2) 157.0(3) 156.9(3)
98.1(2)
95.2(2) 96.2(2)
93.4(2) 92.7(2)
99.4(2)
150.39
162.21
99.15
94.20
95.40
94.55
93.8(2)
96.0(2) 96.0(2)
92.2(2) 92.2(2)
89.0(2)
91.70
92.74
87.22
98.10
92.5(2)
O11-Cu-O21 172.2(2)
Cl1-Cu-Cl2
Cl1-Cu-O11
N4‚‚‚O12
O4‚‚‚O22
136.5(1) 138.0(1)
152.14
98.34
2.925(4)
2.756(4)
a Standard deviations in parentheses.
O34) ) 2.62 Å). A chirality plane is defined by the amide
RC(O)-NR2 group, and the crystals are racemic with an
equal distribution of the RP and SP enantiomers. Most
important is the coordination of the tertiary amide nitrogen
atom N1 at a distance of 2.16 Å to the copper(II) center.
This is only ca. 0.10-0.15 Å longer than the typical copper
to secondary amine contacts in common bpa derivatives14,15
but more than 0.3 Å shorter than the corresponding distances
in 3a8 and 3c (see below). However, the tertiary amide
function is a weak donor, and this results in a rather unusual
feature. The apical triflate anion is only 2.21 Å (Cu-O31)
away from the coordination center. Distances of more than
2.4 Å are common in typical Jahn-Teller distorted octahedral
The geometry of the first coordination sphere of 3c is
interesting. At first sight, the overall geometry may be
(16) (a) Feldhaus, R.; Koppe, R.; Mattes, R. Z. Naturforsch. 1996, 51b,
869. (b) Prins, R.; Birker, P. J. M. W. L.; Haasnoot, J. G.; Verschoor,
G. C.; Reedijk, J. Inorg. Chem. 1985, 24, 4128. (c) Bouayad, A.; Bitit,
N.; Deydier, E.; Menu, M.-J.; Dartiguenave, M.; Dartiguenave, Y.;
Duran, H.; Gorrichon, L.; Simard, M.; Beauchamp, A. L. Polyhedron
1993, 12, 479. (d) Alilou, E. H.; Amadei, E.; Giorgi, M.; Pierrot, M.;
Reglier, M. J. Chem. Soc., Dalton Trans. 1993, 549. (e) Wu, L. W.;
Yamagiwa, Y.; Ino, I.; Sugimoto, K.; Kuroda-Sowa, T.; Kamikawa,
T. Polyhedron 1999, 18, 2047.
(17) (a) Scott, M. J.; Holm, R. H. J. Am. Chem. Soc. 1994, 116, 11357.
(b) Diebold, A.; Kyritsakas, N.; Fischer, J.; Weiss, R. Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun. 1996, 52, 632. (c) Blain, I.; Bruno,
P.; Giorgi, M.; Lojou, E.; Lexa, D.; Reglier, M. Eur. J. Inorg. Chem.
1998, 1297.
(14) Hemmert, C.; Renz, M.; Gornitzka, H.; Meunier, B. J. Chem. Soc.,
Dalton Trans. 1999, 3989.
(15) Mandel, B.; Douglas, B. E. Inorg. Chim. Acta 1989, 155, 55.
(18) Tolman, W. B. Inorg. Chem. 1991, 30, 4877.
4666 Inorganic Chemistry, Vol. 43, No. 15, 2004

Tertiary Carboxamides in Metal Complexes
classified as distorted trigonal-bipyramidal with the two
pyridine donor ligands in the axial positions. However, the
Cu-N(amide) distance (2.46 Å/S,SP; 2.44 Å/S,RP) is longer
than the sum of the van der Waals radius of nitrogen (1.6
Å) and the ionic radius of 5-coordinate Cu2+ (0.79 Å).19
Furthermore, the Cu-N(amide) distance is ca. 0.2 Å longer
than the longest Cu-Cl contact (2.27 Å). It is therefore more
appropriate to describe the structure as monocapped distorted
tetrahedral. Only weak electrostatic interactions are active
between the amide nitrogen atom and the metal ion. This
interpretation is further supported by the theoretical results
described below. It is also consistent with the 2 Cu-Cl
distances in 3c, which differ by only ca. 0.04 Å (2.23 Å/2.27
Å). The similar Cu-Cl bond lengths and the long Cu-N
distance are in marked contrast to observations in most of
the known CuCl2 complexes with substituted bpa lig-
ands.14,15,20,21 These compounds usually have an idealized
basal plane defined by the shortest Cu-ligand contacts and
formed by only one chloro ligand and three bpa nitrogen
atoms. The remaining second Cu-Cl contact is the longest
metal-ligand bond. Depending on their trigonality index22
the geometries can be characterized as square-pyramidal (τ
< 0.2, ∆(Cu-Cl) ) 0.21-0.46 Å) or trigonally distorted
square-pyramidal (τ > 0.2, ∆(Cu-Cl) ) 0.07-0.26 Å). The
only exception is a N-pyridine substituted complex in which
∆(Cu-Cl) ) 0.07 Å and the Cu-N(amine) bond is 0.013
longer than the longest Cu-Cl bond.23 However, this
difference is much smaller than in the tertiary amide
complexes, putting the compound closer to the common bpa
derivatives.
[(bpa)CuCl2] (5). The methanol cleavage of 3a-d finally
resulted in the formation of complex 5. This compound has
been known for quite some time, and considering the
widespread application of bpa derivatives in copper(II)
coordination chemistry, it is somewhat surprising that its
structure has only been reported in December 2003.24 The
literature data show some interesting differences from our
results which are most probably due to the crystallization
conditions. We have obtained crystals from methanol whereas
Choi et al. have used a water/acetonitrile mixture. An ORTEP
plot of our structure is shown in Figure 4. The compound is
square-pyramidal with an equatorial d(Cu-Cl) of 2.23 Å,
and a typical elongated axial d(Cu-Cl) of 2.64 Å. A
trigonality index τ of 0.05 is calculated, close to the ideal
value of 0 for square-pyramidal complexes.22 Choi et al. have
Figure 4. ORTEP plot (30% ellipsoids) of the structure of 5. Selected
bond lengths (Å) and angles are listed in Table 2.
Table 2. Selected Bond Lengths (Å) and Angles (deg) of the bpa
Complexes 5 and 6a
6‚CH3OH
5
A
B
C
Cu-N1
2.007(2)
2.013(2)
2.022(2)
2.233(1)
1.994(8)
1.996(5)
1.996(5)
2.266(2)
1.996(5)
1.989(5)
1.989(5)
2.27(2)
1.987(5)
2.007(4)
2.007(4)
2.219(2)
Cu-N2
Cu-N3
Cu-Cl1
Cu-Cl1a
2.32(2)
Cu-Cl2
2.609(1)
2.658(2)
Cu-O11
2.507(7)
83.2(2)
83.2(2)
166.3(5)
170.5(7)
N1-Cu-N2
N1-Cu-N3
N1-Cu-Cl1
N1-Cu-Cl1a
N1-Cu-Cl2
N1-Cu-O11
N2-Cu-N3
N2-Cu-Cl1
N2-Cu-Cl1a
N2-Cu-Cl2
N2-Cu-O11
N3-Cu-Cl1
N3-Cu-Cl1a
N3-Cu-Cl2
N3-Cu-O11
Cl1-Cu-Cl2
Cl1-Cu-O11
Cl1a-Cu-O11
81.4(1)
81.7(5)
159.6(1)
82.4(2)
82.4(2)
167.4(5)
81.9(2)
81.9(2)
179.0(2)
93.4(1)
88.4(1)
80.8(2)
162.8(1)
97.5(1)
164.4(3)
97.1(2)
165.3(3)
108.9 (3)
106.1(3)
163.3(2)
98.2(2)
90.9(1)
97.1(1)
93.7(1)
107.0(1)
91.4(2)
97.1(2)
91.4(2)
104.2(1)
91.5(2)
108.9 (3)
106.1(3)
98.2(2)
91.5(2)
92.4(4)
101.2(4)
a Standard deviations in parentheses.
reported a slightly more trigonally distorted structure with τ
) 0.12 and correspondingly different Cu-Cl distances of
2.25 and 2.41 Å. Particularly notable is the large effect of
the slight trigonal distortion on the axial bond which dif-
fers by more than 0.2 Å in the two structures. Almost
ideally square-pyramidal geometries were also found in other
CuCl2 complexes of bpa-derived ligands.20,21b However,
structures with trigonally distorted coordination spheres (τsqp
) 0; τtbpy ) 1; τobserved ) 0.18-0.66) are also commonly
observed.14,15,21 The relevant geometric parameters of the
[(bpa)Cu]2+ fragment in 5 are in good agreement with those
in the structurally related complex [(bpa)Cu(CH3OH)]-
(ClO4)2.9a
{[(bpa)CuCl2][(bpa)Cu(H2O)Cl][Cu(bpa)Cl][CuCl4]‚
CH3OH} (6). Compound 6 was obtained as an intermediate
product in the methanolysis of 3a, 3b, and 3d. It contains
three different chloro complexes of the [(bpa)Cu]2+ fragment
and a tetrachlorocuprate(II) dianion. The structure of [(bpa)-
CuCl2] in 6 is identical within the experimental errors to
that of 5. The square-pyramidal complex [(bpa)Cu(H2O)-
Cl]+ and the square-planar complex [Cu(bpa)Cl]+ are shown
in Figure 5. These components have structures similar to
those of related complexes described by Mascharak et al.25
Only the axial Cu-H2O distance of 2.51 Å in our compound
is notably longer than the reported Cu-methanol and
(19) Wiberg, N.; Hollemann, A. F. Lehrbuch der anorganischen Chemie,
101st ed.; W. de Gruyter: Berlin, 1995; p 1838.
(20) (a) Yoon, D.-C.; Oh, C.-E.; Lee, U. Acta Crystallogr., Sect. E: Struct.
Rep. Online 2002, 58, m113. (b) Evans, A. J.; Watkins, S. E.; Craig,
D. C.; Colbran, S. B. J. Chem. Soc., Dalton Trans. 2002, 983. (c) He,
Z.; Colbran, S. B.; Craig, D. C. Chem. Eur. J. 2003, 9, 116.
(21) (a) Nagao, H.; Komeda, N.; Mukaida, M.; Suzuki, M.; Tanaka K.
Inorg. Chem. 1996, 35, 6809. (b) Kani, Y.; Ohba, S.; Ito, S.; Nishida
Y. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 2000, 56, e195.
(c) Ugozzoli, F.; Massera, C.; Manotti-Lanfredi, A. M.; Marsich, N.;
Camus, A. Inorg. Chim. Acta 2002, 340, 97.
(22) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G.
C. J. Chem. Soc., Dalton Trans. 1984, 1349.
(23) Foxon, S. P.; Walter, O.; Schindler, S. Eur. J. Inorg. Chem. 2002,
111.
(24) Choi, K.-Y.; Ryu, H.; Sung, N.-D.; Suh, M. J. Chem. Crystallogr.
2003, 33, 947.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4667

Niklas et al.
Figure 5. ORTEP plots (30% ellipsoids) of the structures of the
components [(bpa)CuCl2] (A), [(bpa)Cu(H2O)Cl]+ (B), and [(bpa)CuCl]+
(C) in 6. Selected bond lengths and angles are listed in Table 2.
Figure 7. Ball and stick plot of the calculated structure of [(Ac-bpa)-
CuCl(CH3OH)]+. Selected bond lengths and angles are listed in Table 1.
by Reedijk et al. for a sterically hindered copper complex
of the hexadentate EDTB (EDTB ) N,N,N′,N′-tetrakis[(2-
benzimidazolyl)methyl]-1,2-ethanediamine).28
[(Ac-bpa)CuCl(CH3OH)]+. The X-ray structural analysis
of the mixed salt 6 revealed that monochloro complexes may
be involved in the methanol cleavage reactions of 3a-d.
We reasoned that they could be the actual reactive species,
since the dichloro complexes show only a rather small
activation of their amide functions. Our aim was to compare
the calculated structure of an N-bound tertiary amide in the
monochloro complex [(Ac-bpa)CuCl(CH3OH)]+ with the
structure of the strongly activated triflate salt 2b. It turned
out that the two structures are strikingly similar. An ORTEP
representation of [(Ac-bpa)CuCl(CH3OH)]+ is shown in
Figure 7. The coordination geometry is square-pyramidal
with Cl-, the two pyridine donors, and the tertiary amide
nitrogen atom in the basal plane. Most important, the Cu-
Namide distance of 2.22 Å is only slightly longer than that in
2b (2.164 Å). The Cu-Cl bond length is 2.24 Å. The
methanol is weakly bound at a distance of 2.363 Å in the
apical position. It is interesting to note the close relationship
between [(Ac-bpa)CuCl(CH3OH)]+ and 2b, and the cleavage
Figure 6. Ball and stick plot of the calculated structure of [(Ac-bpa)-
CuCl2]. Selected bond lengths and angles are listed in Table 1.
-ethanol contacts of 2.33 and 2.32 Å, respectively. The
tetrachlorocuprate(II) ion has a typical distorted tetrahedral
structure26 and is disordered about a crystallographic mirror
plane. It should be noted that the [(bpa)Cu(H2O)Cl]+ cation
and the solvent molecule in 6 are also severely disordered.
This has prompted us to confirm the constitution not only
by C,H,N-elemental analysis, but also by a determination
of the crystal density. These data confirm the structural
assignments. The importance of the structure of 6 lies not
so much in the geometric details but rather in the fact that it
indicates the presence of monochloro complexes and free
chlorocopper(II) complexes in the reaction mixtures.
Calculated Structures
9a
products 5, [(bpa)CuCl]+ in 6 and [(bpa)Cu(H2O)](ClO4)2
which have similar coordination geometries. Only a shorten-
ing of one Cu-N distance from ca. 2.2 (Cu-Namide) to ca.
2.0 Å (Cu-Namine) is required when [(bpa)CuL] acts as a
leaving group.
[(Ac-bpa)CuCl2]. Since this study represents the first
attempt to model N-bound tertiary amide complexes, our first
aim was to validate the calculations by comparison of
theoretical and crystallographic results. In order to reduce
the number of geometric parameters, the acyl substituents
in 1a-d were replaced by an acetyl group in the model
ligand Ac-bpa. The first calculated structure is that of the
complex [(Ac-bpa)CuCl2]. The result is shown in Figure 6.
Remarkably good agreement with the experimental structures
of 3a, 3c, and 3d is evident. The longest distance is found
between the Cu(II) center and the amide-nitrogen atom. With
2.627 Å it is clearly outside the range of a covalent
interaction. Consequently, the flattened tetrahedral N2Cl2-
geometry is even more pronounced than that in 3c. This
Jahn-Teller distorted geometry is frequently observed in
tetracoordinate copper(II) complexes containing two nitrogen
and two chloro ligands27 and is consistent with a “2 + 1”
binding mode of the acylated bpa-ligand. This coordination
mode is rare in copper(II) complexes with tri- and polyden-
tate amine ligands. A comparable situation was described
Activation of the Amide Group. In order to understand
the C-N bond solvolysis, it is necessary to consider the
activation of the amide group in the complexes 2a-d and
3a-d, as well as in the calculated structures of [(Ac-bpa)-
CuCl2] and [(Ac-bpa)CuCl(CH3OH)]. Figure 8 contains an
illustration of the first coordination spheres. It shows that
the dichloro complexes are characterized by a weak, non-
covalent copper to amide nitrogen contact (Figure 8a). Their
monocapped distorted tetrahedral N2Cl2 coordination sphere
is distinctly different from the distorted trigonal-bipyramidal
geometry in related bpa complexes without a tertiary
carboxamide unit (Figure 8b).14,15,21 It is interesting to
compare our dichloro complexes with copper complexes of
(27) Selected examples: (a) Brown, D. B.; Hall, J. W.; Helis, H. M.;
Walton, E. G.; Hodgson, D. J.; Hatfield, W. E. Inorg. Chem. 1977,
16, 2675. (b) Cingi, M. B.; Manotti Lanfredi, A. M.; Tiripicchio, A.
Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1983, 39, 1523.
(c) van Ooijen, J. A. C.; Reedijk, J.; Spek, A. L. J. Chem. Soc., Dalton
Trans. 1979, 1183.
(25) Rowland, J. M.; Olmstead, M. M.; Mascharak, P. K. Inorg. Chem.
2000, 39, 5326.
(26) van Berkel, P. M.; Driessen, W. L.; Ha¨ma¨la¨inen, R.; Reedijk, J.;
Turpeinen, U. Inorg. Chem. 1994, 33, 5920.
(28) Birker, P. J. M. W. L.; Hendriks, H. M. J.; Reedijk, J.; Verschoor, G.
C. Inorg. Chem. 1981, 20, 2408.
4668 Inorganic Chemistry, Vol. 43, No. 15, 2004

Tertiary Carboxamides in Metal Complexes
Figure 9. Distortion of the formally planar amide function in the calculated
structures of [(Ac-bpa)CuCl2] and [(Ac-bpa)CuCl(CH3OH)], as well as in
the crystal structures of 2b and 3c.
C(O)R groups in the two calculated and two experimental
structures. The agreement between experiment and theory
is very good. It is evident that the dichloro-complexes (øN
) ca. 30°) are significantly less distorted than the monochlo-
ro- and triflato-derivatives (øN ) ca. 45°). Concomitant with
a higher degree of sp3 pyramidalization and loss of resonance
stabilization is a lengthening of the C-N bond from ca. 1.39
to ca. 1.44 Å. For comparison, the free ligands 1c and 1d
have shorter C-N distances of ca. 1.35 Å and øN-values of
10.4° and 8.2°.9 These data are typical for unperturbed
tertiary amide groups.
Figure 8. First coordination sphere of (a) the monocapped distorted-
tetrahedral complexes 3c (X-ray), [(Ac-bpa)CuCl2] (calculated), (b) the
distorted trigonal-bipyramidal complex N-(2-aminoethyl)bis(2-picolyl)-
amino-dichlorocopper(II)-hydroperchlorate,15 and (c) the square-pyramidal
complexes 2b (X-ray) and [(Ac-bpa)CuCl(CH3OH)] (calculated).
The C-O bond distances are less useful for a quantitative
evaluation of structural effects than the C-N distances.32
However, the carbonyl stretching frequencies are highly
sensitive measures for the degree of resonance in an amide
function. This has been extensively exploited in studies on
purely organic distorted amides which show nice correlations
between the C-N bond lengths and the CO stretching
vibration.33 In metal complexes, N-coordination results in
higher, and O-coordination in lower, frequencies compared
to the free amide ligand. For the dichloro-complex 3c, we
observe a value of 1690 cm-1 whereas the triflato-complex
2b exhibits a band at 1756 cm-1. The free ligand resonances
are at 1674 (1c) and 1660 (1b) cm-1, respectively. Thus,
the øN values correspond to the magnitude of the high
frequency shift which is most pronounced in the highly
distorted derivative 2b. The complexes for which no crystal-
lographic information is available clearly fall into the same
ranges, confirming the similarity of their structures. It is
interesting to compare our data with those of the most twisted
organic amide, a 1-aza-2-adamantanone reported by Kirby
et al.34 This compound has a C-N bond length of 1.475 Å
and an IR-stretching frequency of 1732 cm-1. The authors
have shown that it behaves like an aminoketone rather than
an amide. This is enforced by a large twist angle τ of ca.
90° which completely removes the amide resonance. The
twist angle τ is the angle between the C(N)-C(N) axis and
the O(C)-C(C) axis and ranges from 0° in a planar amide
a monoacylated triazacylononane ligand which was recently
reported by Houser et al.29 The tertiary amide function in
these compounds does not bind to the metal center, and the
triazacycononane binds in a bidentate ethylenediamine
fashion. Due to the ligand geometry, the amide nitrogen
points away from the metal center and is not activated.
Schindler et al. have reported an amide substituted copper
complex of bpa which contains the neutral nitrogen atom in
the apical position of a square pyramid, 2.64 Å away from
the central copper ion.30 The authors reported a small
deplanarization of the amide moiety as a consequence of
weak electrostatic interactions, but the effect is much smaller
than in our complexes. Our ligands keep the nitrogen atom
in a position that enforces electrostatic interactions. As a
consequence, a significant pyramidalization of the amide
group is observed in 3a-d although less pronounced than
in the structures of the square pyramidal triflato complex
2b and the monochloro complex [(Ac-bpa)CuCl(CH3OH]+,
which contain a strongly bound tertiary amide function
(Figure 8c).
The geometry of the amide moiety is best described by
the pyramidalization øN at the nitrogen atom.31 The parameter
øN reflects the degree of sp3 hybridization as a consequence
of metal-ion coordination. It is conveniently calculated from
the C-N-C-O and C-N-C-C torsion angles of the amide
group and increases from 0° (sp2 hybridization) to 60° (sp3
hybridization) when N is completely pyramidal. This cor-
responds to a significant activation of the amide function
by loss of resonance stabilization. Figure 9 shows the R2N-
(32) Yamada, S. In The Amide Linkage: Selected Structural Aspects in
Chemistry, Biochemistry, and Materials Science; Greenberg, A.,
Breneman, C. M., Liebman, J. F., Eds.; J. Wiley & Sons: Weinheim,
2000; p 215.
(33) Greenberg, A.; Chiu, Y.-Y.; Johnson, J. L.; Liebman, J. F. Struct.
Chem. 1991, 2, 117.
(34) (a) Kirby, A. J.; Komarov, I. V.; Wothers, P. D.; Feeder, N. Angew.
Chem., Int. Ed. Engl. 1998, 37, 785. (b) Kirby, A. J.; Komarov, I. V.;
Feeder, N. J. Am. Chem. Soc. 1998, 120, 7101. (c) Kirby, A. J.;
Komarov, I. V.; Feeder, N. J. Chem. Soc., Perkin Trans. 2 2001, 522.
(29) Mondal, A.; Klein, E. L.; Khan, M. A.; Houser, R. P. Inorg. Chem.
2003, 42, 5462.
(30) Schatz, M.; Leibold, M.; Foxon, S. P.; Weitzer, M.; Heinemann, F.
W.; Hampel, F.; Walter, O.; Schindler, S. Dalton Trans. 2003, 1480.
(31) Winkler, F. K.; Dunitz, J. D. J. Mol. Biol. 1971, 59, 169.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4669

Niklas et al.
1G spectrophotometer. IR (KBr): Mattson Polaris FTIR. Elemental
analysis: Carlo Erba elemental analyzer model 1106. FD- and FAB-
MS: Varian MAT 212.
to 90° in a completely distorted amide function. Generally,
a geometrically enforced large twist angle is the means of
activation in purely organic carboxamides.32 We have shown
that N-coordination of a metal ion to a tertiary amide has
the same activating effect but does not require any steric
strain in the amide. The significantly enhanced nucleophilic
cleavage lability of the C-N bond is common to both
Werner-type complexes and twisted amides. However,
whereas numerous experimental35 and theoretical36 studies
have dealt with the C-N-bond hydrolysis in organic
compounds, we are only at the beginning of understanding
the corresponding reactivity of metal complexes. A major
obstacle is the inherent lability of N-coordinated tertiary
carboxamides. We have shown earlier that the ligands 1a-d
do not form stable copper(II) complexes in aqueous solutions
at pH values below ca. 937 and that addition of propylamine
is sufficient to extract the metal ion from acylated bpa
ligands.9a The formation of tetrachlorocuprates in the present
study provides evidence for a dissociation in methanol
solutions. Theoretical studies are therefore important to gain
insight into the actual reactive species in solution. The first
results are presented in this paper. We were able to show
that the formation of monochloro complexes LCuCl+ nicely
explains the similar reactivity of triflato and chloro com-
plexes in methanolysis reactions.
All complexes were prepared and stored under an atmosphere
of dry nitrogen. Absolute solvents were purchased from Fluka and
stored under nitrogen. Solvents were used without further purifica-
tion. The water used for extractions was bidistilled. All other
reagents were of commercially available reagent grade quality.
Amino acid derivatives were purchased from Bachem and copper-
(II) triflate from Acros. All other chemicals were from Aldrich.
Bis[(2-pyridyl)methyl]amine (bpa) was prepared according to
literature procedures.38
Chromatographic separations were achieved on flash columns
under nitrogen pressure. The stationary phase was silica (Merck
Type 9385, 230-400 mesh, 60 Å) from Aldrich. CH3OH/CH2Cl2
mixtures were used as eluents. Technical grade dichloromethane
was used which was purified by rotary evaporation. Methanol was
of p.a. quality. The separation was optimized and followed by TLC
(silica).
The synthesis of the ligand Boc-Ala-bpa (1c) and its methanolysis
in the presence of Cu(CF3SO3)2, as well as the synthesis of complex
2c, were published previously.9a The ligands 1a,8 1b, and 1d9b were
reported in short communications but without experimental details.
Their syntheses are described below.
Ligands. A general method was used for the DCC/HOBt
coupling, but the following workup conditions differed significantly
for each compound.
DCC/HOBt Coupling (General Method). A THF solution
containing equimolar amounts of the carboxylic acid component,
bpa, and 1.2 equiv of hydroxybenzotriazol (HOBt) was cooled to
-10 °C in an ice/salt bath. N,N′-Dicyclohexylcarbodiimid (DCC,
1.2 equiv) was dissolved in a small volume of THF and added in
one portion to the solution. The mixture was stirred at -10 °C
over a period of 1 h and then allowed to warm to room temperature.
Stirring was continued overnight. The suspension was filtered off
and washed several times with small amounts of cold (4 °C) THF.
The combined filtrates were evaporated to dryness. The following
workup is described separately for each compound.
Conclusions
The activation of tertiary carboxamides by coordination
of their electrically neutral nitrogen atom to a metal center
was postulated 30 years ago as a means of promoting the
nucleophilic C-N bond cleavage in acylated [(R-C(O)bpa)-
Cu]2+ complexes.6 Our present paper provides the first
detailed structural investigation of such complexes and their
reaction products. Theoretical studies were performed to gain
a better understanding of the reactive species in solution.
The importance of ligand dissociation and exchange pro-
cesses in reactions of N-bound tertiary carboxamides is
evident. We were able to show that the coordination
properties of a neutral amide nitrogen atom can range from
covalent binding with strong activation to weak electrostatic
interactions. This depends on the co-ligands, triflate and
chloride in our study. Our results further suggest that metal
coordination may have similar effects on an amide function
to steric strain in purely organic compounds. With our work,
we are beginning to answer what are perhaps some of the
last open questions in the coordination chemistry of the amide
group.
PhC(O)-bpa (1a). The amounts of each reagent follow: benzoic
acid (5.00 g, 40.9 mmol), bpa (8.15 g, 40.9 mmol), HOBt (6.65 g,
49.1 mmol), DCC (10.14 g, 49.1 mmol), THF (50 mL). The crude
product was dissolved in CH2Cl2 (100 mL) and washed 3 times
with 100 mL of a 0.5 M NaHCO3-solution, and finally washed
with 100 mL of water. The organic phase was dried over MgSO4,
filtered and rotary-evaporated to dryness. Flash column chroma-
tography using 20:1 CH2Cl2/CH3OH as the eluent yielded the ligand
as the third fraction (Rf ) 0.45). Compound 1a was isolated as a
light yellow solid which was recrystallized from warm (C2H5)2O
in an ultrasound bath. Colorless needles precipitated after a short
period of standing at room temperature; these were filtered off and
dried under vacuum (9.25 g, 30.5 mmol, 75%).
MS (FD, CHCl3): m/z ) 608 [2M]+, 304 [M]+. IR (KBr,
1
Experimental Section
[cm-1]): ν ) 1634 (νCdO, amide), 1590 (νCdN). H NMR (300
MHz, CDCl3, [ppm]): δ ) 4.69 (s, 2H; py-CH2), 4.89 (s, 2H; py-
CH2), 7.19 (m, 3H; 2 × H5-py, H4-Ph), 7.32-7.45 (m, 4H; 2 ×
H3-py, H3-Ph, H5-Ph), 7.56 (m, 2H; H2-Ph, H6-Ph), 7.67 (m, 2H;
2 × H4-py), 8.55 (m, 2H; 2 × H6-py).
General Methods. Spectra were recorded with the following
instruments. H NMR: Bruker Avance DPX 300. All chemical
shifts are referenced to TMS as internal standard, with high-
frequency shifts recorded as positive. UV-vis spectra: Varian Cary
1
Boc-Gly bpa (1b). The amounts of each reagent follow: Boc-
Gly-OH (5.00 g, 28.5 mmol), bpa (5.68 g, 28.5 mmol), HOBt (4.63
(35) Review: Brown, R. S. In The Amide Linkage: Selected Structural
Aspects in Chemistry, Biochemistry, and Materials Science; Greenberg,
A., Breneman, C. M., Liebman, J. F., Eds.; J. Wiley & Sons:
Weinheim, 2000; pp 85 and references therein.
(36) Lopez, X.; Mujika, J. I.; Blackburn, G. M.; Karplus, M. J. Phys. Chem.
A 2003, 107, 2304 and references therein.
(37) Geisser, B.; Alsfasser, R. J. Chem. Soc., Dalton Trans. 2003, 612.
(38) (a) Hojland, F.; Toftlund, H.; Yde-Andersen, S. Acta Chem. Scand.
1983, A37, 251. (b) Romary, J. K.; Zachariasen, R. D.; Barger, J. D.;
Schiesser, H. J. Chem. Soc. C 1968, 2884.
4670 Inorganic Chemistry, Vol. 43, No. 15, 2004

Tertiary Carboxamides in Metal Complexes
g, 34.4 mmol), DCC (7.07 g, 34.4 mmol), THF (100 mL). The
crude product was dissolved in 100 mL of CH2Cl2 (100 mL) and
washed 3 times with citric acid (50 mL, 0.05 M), 1 time with H2O
(50 mL), 3 times with NaHCO3 (0.5 M, 50 mL), and 1 time with
H2O (100 mL). The organic phase was dried over MgSO4, filtered,
and rotary-evaporated to dryness. Purification was achieved by flash
column chromatography using 15:1 CH2Cl2/CH3OH as the eluent.
Several small yellow and brown fractions eluted first which were
followed by ligand 1b in a broad, light-yellow band (Rf ) 0.5).
Rotary-evaporation of the solvent and drying under vacuum yielded
1b as a light-yellow, viscous oil (6.89 g, 19.3 mmol, 68%).
MS (FD, CHCl3): m/z ) 714 [2M]+, 357 [M]+. IR (KBr, film,
[cm-1]): ν ) 3421 (νNsH), 3320 (νNsH), 1711 (νCdO, urethane),
[Cu(Boc-Gly-bpa)(H2O)(CF3SO3)2] (2b). The amounts of each
reagent follow: Boc-Gly bpa (1b) (863 mg, 2.42 mmol), Cu(CF3-
SO3)2 (875 mg, 2.42 mmol), CH3CN (20 mL). The blue crude
product was treated with CH2Cl2 until it was mostly dissolved. The
green solution was filtered and left for crystallization. A microc-
rystalline solid precipitated which was filtered off and washed with
CH2Cl2 until the filtrate remained colorless. This procedure afforded
2b as a pure, dark-blue solid which was readily soluble in CH3CN
but not longer in CH2Cl2 (1.201 g, 1.63 mmol, 67%). X-ray quality
crystals were obtained from a highly dilute CH2Cl2 solution. Blue
needles of 2b‚CH2Cl2 crystallized upon standing at RT for several
days.
Anal. (%) Calcd for C21CuF6H26N4O10S2 (736.12 g/mol): C
34.26, H 3.56, N 7.61. Found: C 34.13, H 3.54, N 7.46. MS (FAB,
NBA): m/z ) 568 [Cu(Boc-Gly-bpa)(CF3SO3)]+, 512 [(Cu(Boc-
Gly-bpa)(CF3SO3)) - H2CdC(CH3)2]+, 419 [Cu(Boc-Gly-bpa)]+,
319 [(Cu(Boc-Gly bpa)) - (CH3)3CCOO]+. IR (KBr, [cm-1]): ν
) 3414 (br, νOsH, H2O), 1756 (νCdO, amide), 1701 (νCdO, urethane),
1667 (sh), 1615 (νCdN), 1289 (νSO ), 1244 (νCF ), 1167 (νCF ), 1032
1
1660 (νCdO, amide), 1592 (νCdN). H NMR (300 MHz, CDCl3,
3
[ppm]): δ ) 1.43 (s, 9H; C(CH3)3), 4.18 (d, J(RCH2,NH) ) 5.5
R
Hz, 2H; CH2), 4.64 (s, 2H; py-CH2), 4.77 (s, 2H; py-CH2), 5.59
(s, br, 1H; NH), 7.15-7.29 (m, 4H; 2 × H3-py, 2 × H5-py), 7.64
3
(m, 2H; 2 × H4-py), 8.50 (d, J(H6,H5) ) 6.5 Hz, 1H; H6-py),
3
8.56 (d, J(H6,H5) ) 4.5 Hz, 1H; H6-py).
3
3
3
(νSO ). UV-vis (CH3CN, λ [nm] (ꢀ)): 648 (62).
Boc-Leu-bpa (1d). The amounts of each reagent follow: Boc-
Leu-OH (5.00 g, 20.1 mmol), bpa (3.99 g, 20.1 mmol), HOBt (3.25
g, 24.1 mmol), DCC (4.97 g, 24.1 mmol), THF (100 mL). The
crude product was dissolved in CH2Cl2 (100 mL) and washed 3
times with citric acid (100 mL, 0.05 M), 1 time with saturated NaCl
(100 mL), 3 times with NaHCO3 (0.5 M, 100 mL), and 1 time
with saturated NaCl (100 mL). The organic phase was dried over
MgSO4 and filtered and the volume reduced to ca. one-third.
Residual DCH was filtered off, the solvent removed by rotary
evaporation and the residual dried under vacuum. The crude product
was recrystallized several times from (C2H5)2O. This was achieved
by dissolving the solid in large amounts of ether with heating and
subsequent storage at -20 °C. The resulting colorless product 1d
was isolated by filtration and dried under vacuum (4.47 g, 10.8
mmol, 54%).
3
[Cu(Boc-Leu-bpa)(H2O)(CF3SO3)2] (2d). The amounts of each
reagent follow: Boc-Leu-bpa (1d) (349 mg, 0.85 mmol), Cu(CF3-
SO3)2 (307 mg, 0.85 mmol), CH3CN (10 mL). The crude product
was dissoved in 10 mL of CH2Cl2. Slow diffusion of (C2H5)2O to
this solution resulted in separation of a blue oil. The mother liquor
was decanted off and the oil dried under vacuum. This afforded
pure 2d as a dark-blue to turquoise solid foam. (506 mg, 0.64 mmol,
75%).
Anal. (%) Calcd for C25CuF6H34N4O10S2 (792.23 g/mol): C
37.87, H 4.33, N 7.07. Found: C 37.90, H 4.66, N 7.00. MS (FAB,
NBA): m/z ) 624 [Cu(Boc-Leu-bpa)(CF3SO3)]+, 475 [Cu(Boc-
Leu-bpa)]+, 375 [(Cu(Boc-Leu-bpa)) - (CH3)3CCOO]+. IR (KBr,
[cm-1]): ν ) 3420 (br, νOsH, H2O), 1747 (νCdO, amide), 1698 (sh),
1660 (νCdO, urethane), 1614 (νCdN), 1283 (νSO ), 1252 (νCF ), 1166
3
3
MS (FD, CHCl3): m/z ) 826 [2M]+, 413 [M]+. IR (KBr,
(νCF ), 1031 (νSO ). UV-vis (CH3CN, λ [nm] (ꢀ)): 664 (66).
3
3
[cm-1]): ν ) 3384 (νNsH), 3180 (νNsH), 1700 (νCdO, urethane),
[LCuCl2] Complexes: General Method. Because of their light
sensitivity the dichloro copper(II) complexes 3a-d were synthe-
sized and stored in the dark. Equimolar quantities of CuCl2‚2H2O
and the respective ligand were reacted in acetonitrile.
[Cu(PhC(O)-bpa)(Cl)2] (3a). The amounts of each reagent
follow: Benz-bpa (1a) (704 mg, 2.32 mmol), CuCl2‚2H2O (396
mg, 2.32 mmol), CH3CN (15 mL). Precipitation of the product from
the reaction mixture started after several minutes of stirring. The
suspension was stirred overnight and thereafter stored for 24 h at
-20 °C. Filtration, repeated washing with small amounts of
acetonitrile, and drying under vacuum afforded 3a as a turquoise
powder (825 mg, 1.88 mmol, 81%).
Anal. (%) Calcd for C19Cl2CuH17N3O (437.81 g/mol): C 52.13,
H 3.91, N 9.60. Found: C 52.57, H 3.99, N 9.48. MS (FAB,
NBA): m/z ) 401 [Cu(PhC(O)-bpa)(Cl)]+, 366 [Cu(PhC(O)-bpa)]+,
304 [PhC(O)-bpa]+. IR (KBr, [cm-1]): ν ) 1673 (νCdO, amide),
1608 (νCdN). UV-vis (CH2Cl2, λ [nm] (ꢀ)): 789 (293).
[Cu(Boc-Gly-bpa)(Cl)2] (3b). The amounts of each reagent
follow: Boc-Gly bpa (1b) (605 mg, 1.70 mmol), CuCl2‚2H2O (290
mg, 1.70 mmol), CH3CN (15 mL). A light blue solid precipitated
after several minutes. Stirring was continued overnight and the
resulting suspension stored for 24 h at -20 °C. Product 3b was
filtered off and dried under vacuum to yield a light-blue powder
(714 mg, 1.45 mmol, 85%).
1
1640 (νCdO, amide), 1593 (νCdN). H NMR (300 MHz, CDCl3,
3
γ
3
[ppm]): δ ) 0.80, 0.86 (2 × d, J(CH3, CH) ) 6.4 Hz, J(CH3,
γCH) ) 6.6 Hz, 6H; CH(CH3)2), 1.43-1.57 (m, 2H; CH2), 1.43
γ
â
γ
(s, 9H; C(CH3)3), 1.61-1.70 (m, 1H; CH), 4.64-4.76 (m, 3H; 2
× py-CHAHB, RCH), 4.85 (d, J(CHA, HB) ) 15.4 Hz, 1H; py-
2
CHAHB), 5.50 (d, J(CHA, HB) ) 17.4 Hz, 1H; py-CHAHB), 5.19
2
(d, 3J(NH, CHR) ) 9.0 Hz, 1H; NH), 7.14-7.28 (m, 4H; 2 × H3-
3
py, 2 × H5-py), 7.63 (m, 2H; 2 × H4-py), 8.50 (d, J(H6, H5) )
3
4.2 Hz, 1H; H6-py), 8.56 (d, J(H6, H5) ) 4.6 Hz, 1H; H6-py).
[LCu(CF3SO3)2]-Complexes: General Method. An equimolar
amount of Cu(CF3SO3)2 was added to an acetonitrile solution of
the ligand. The resulting dark-blue solution was stirred overnight
and all solvent removed under vacuum.
[Cu(PhC(O)-bpa)(CF3SO3)2] (2a). The amounts of each reagent
follow: Benz-bpa (1a) (301 mg, 0.99 mmol), Cu(CF3SO3)2 (358
mg, 0.99 mmol), CH3CN (10 mL). The crude product was dissolved
in 10 mL of CH2Cl2. Within several minutes, the solid product 2a
precipitated. The resulting suspension was left overnight at 4 °C.
After filtration and drying under vacuum, pure 2a was obtained as
a turquoise powder (606 mg, 0.91 mmol, 92%).
Anal. (%) Calcd for C21CuF6H17N3O7S2 (665.04 g/mol): C 37.93,
H 2.58, N 6.32. Found: C 37.75, H 2.71, N 6.13. MS (FAB,
NBA): m/z ) 515 [Cu(PhC(O)-bpa)(CF3SO3)]+, 366 [Cu(PhC(O)-
bpa)]+, 304 [PhC(O)-bpa]+. IR (KBr, [cm-1]): ν ) 1701 (νCdO
,
Anal. (%) Calcd for C19Cl2CuH24N4O3 (490.87 g/mol): C 46.49,
H 4.93, N 11.41. Found: C 46.80, H 5.12, N 11.31. MS (FAB,
NBA): m/z ) 454 [Cu(Boc-Gly-bpa)(Cl)]+, 419 [Cu(Boc-Gly-
bpa)]+, 356 [Boc-Gly-bpa]+. IR (KBr, [cm-1]): ν ) 3441 (νNsH),
amide), 1678 (sh), 1615 (νCdN), 1318 (νSO ), 1281 (νSO ), 1246
3
3
(νCF ), 1230 (νCF ), 1165 (νCF ), 1032 (νSO ). UV-vis (CH3CN, λ
3
3
3
3
[nm] (ꢀ)): 632 (58).
Inorganic Chemistry, Vol. 43, No. 15, 2004 4671

Niklas et al.
Table 3. Crystallographic Data for 2b, 3c, 5, 5‚2H2O, and 6
2b‚CH2Cl2
3c‚0.5H2O‚CH2Cl2
5
5‚2H2O
6‚CH3OH
formula
fw
C22H28Cl2CuF6N4O10S2
821.04
C21H29Cl4CuN4O3.50
598.82
C12H13Cl2CuN3
336.69
C12H17Cl2CuN3O2
369.73
C37H45Cl18Cu4N9O2
1185.58
cryst size [mm3]
T [K]
0.96 × 0.28 × 0.12
0.40 × 0.30 × 0.10
0.29 × 0.29 × 0.17
0.23 × 0.09 × 0.07
0.29 × 0.29 × 24
200(2)
173(2)
100(2)
100(2)
100(2)
cryst syst
space group
a [Å]
b [Å]
c [Å]
monoclinic
P21/c
12.296(1)
15.165(3)
18.470(3)
90
104.83(1)
90
3329.4(9)
4
1.638
1.031
1.76-27.01
8915
7260 (Rint ) 0.0452)
4082
434
monoclinic
P21
16.9604(3)
13.2836(2)
26.5637(5)
90
101.6030(10)
90
5862.37(18)
8
1.357
1.139
2.19-24.11
18137
18137 (Rint ) 0.0000)
13565
1192
monoclinic
P21/n
6.5320(1)
13.2558(2)
15.4821(2)
90
98.985(2)
90
1324.10(3)
4
1.674
2.036
3.35-34.00
33908
5390 (Rint ) 0.0580)
3887
202
orthorhombic
Pbca
15.6476(3)
6.5115(2)
30.6818(6)
90
90
90
3126.2(2)
8
1.571
1.742
3.45-28.00
20895
3780 (Rint ) 0.1221)
1864
181
orthorhombic
Pnma
22.3994(4)
15.3804(3)
13.2115(2)
90
90
90
4551.5(2)
4
1.730 (Dobsd ) 1.735)
2.359
3.82-27.10
33601
5149 (Rint ) 0.0487)
4296
363
1.162
R [deg]
â [deg]
γ [deg]
V [Å3]
Z
D
calcd [Mg/m3]
µ [mm-1
]
θ-range [deg]
measured reflns
indep reflns
obsd reflnsa
params
GOF on F2
R-valuea
1.022
1.350
1.030
0.942
R1 ) 0.0519
wR2 ) 0.1381
0.486/-0.533
R1 ) 0.0708
wR2 ) 0.2029
1.385/-0.703
R1 ) 0.0325
wR2 ) 0.0717
0.471/-0.611
R1 ) 0.0489
wR2 ) 0.0996
455/-812
R1 ) 0.0616
wR2 ) 0.1269
1.041/-0.905
R-value (all data)
max/min [e Å-3
]
a [I > 2σ(I)].
3344 (ν ), 1722 + 1674 (νCdO, amide + urethane), 1607 (νCd
N). UV-vis (CH2Cl2, λ [nm] (ꢀ)): 771 (281).
NBA): m/z ) 510 [Cu(Boc-Leu-bpa)(Cl)]+, 475 [Cu(Boc-Leu-
bpa)]+, 454 [(Cu(Boc-Leu-bpa)(Cl)) - H2CdC(CH3)2]+, 413 [Boc-
Leu-bpa]+, 375 [(Cu(Boc-Leu-bpa)) - (CH3)3CCOO]+. IR (KBr,
[cm-1]): ν ) 3450 (νNsH), 3389 (νNsH), 1682 (νCdO, amide +
urethane), 1610 (νCdN). UV-vis (CH2Cl2, λ [nm] (ꢀ)): 807 (303).
Methanolysis of 1d in the Presence of CuCl2‚H2O. [Cu(bpa)-
Cl2] (5). Addition of CuCl2‚2H2O (77 mg, 0.45 mmol) to a solution
of Boc-Leu-bpa (1d) (187 mg, 0.45 mmol) in CH3OH (5 mL) was
followed by immediate precipitation of [Cu(Boc-Leu-bpa)(Cl)2]
(3d). The precipitate redissolved upon addition of methanol (ca.
15 mL). The resulting turquoise solution was stirred in the dark at
room temperature for 3 days. After reduction of the volume to ca.
one-half and storage of the solution at -20 °C for several days,
product 5 precipitated as a dark-blue solid which was isolated and
dried under vacuum (81 mg, 0.24 mmol, 53%). Slow diffusion of
(C2H5)2O to the reaction mixture at room temperature yielded dark-
blue prisms which were suitable for an X-ray structure analysis.
The compound was also crystallized by slow evaporation of water
from an aqueous solution. This modification yielded a structure
with different crystal parameters due to the presence of two
noncoordinating solvent H2O molecules in the unit cell.
Anal. (%) Calcd for C12Cl2Cu H13N3 (331.71 g/mol): C 43.19,
H 3.93, N 12.59. Found: C 43.33, H 3.97, N 12.43. MS (FAB,
NBA): m/z ) 631 [Cu2(bpa)2Cl3]+, 297 [Cu(bpa)Cl]+, 262 [Cu-
(bpa)]+. IR (KBr, [cm-1]): ν ) 3070 (νNsH), 1606 (νCdN). UV-
vis (CH3OH, λ [nm] (ꢀ)): 673 (118).
[Cu(bpa)Cl2][Cu(bpa)(H2O)Cl][Cu(bpa)Cl][CuCl4]‚CH3OH
(6). CuCl2‚2H2O (148 mg, 0.87 mmol) was added to a solution of
Boc-Leu-bpa (1d) (358 mg, 0.87 mmol) in CH3OH (10 mL). The
precipitated [Cu(Boc-Leu-bpa)(Cl)2] (3d) redissolved upon addition
of CH2Cl2 (ca. 10 mL), and the solution stirred for several hours.
A sample of ca. 5 mL was separated and set for slow diffusion of
(C2H5)2O. Dark-green X-ray quality crystals of 6 were obtained
within 24 h. The compound was insoluble in CH3CN, CH2Cl2, and
CH3OH which prevented its investigation in solution. Drying of
the compound under vacuum resulted in quantitative loss of
methanol from the crystals. Anal. (%) Calcd for C36Cl8Cu4H41N9O
(1153.57 g/mol): C 37.48, H 3.58, N 10.93. Found: C 37.53, H
NsH
[Cu(Boc-Ala-bpa)(Cl)2] (3c). The amounts of each reagent
follow: Boc-Ala-bpa (1c) (387 mg, 1.04 mmol), CuCl2‚2H2O (179
mg, 1.04 mmol), CH3CN (10 mL). The reaction solution was stirred
overnight. The solvent was removed and the remaining crude
product dissolved in CH2Cl2. Slow diffusion of ether into this
solution resulted in the precipitation of 3c as a microcrystalline,
blue solid within several days. The product was filtered off and
dried under vacuum (357 mg; 0.71 mmol, 68%). Crystals for an
X-ray structure analysis were obtained when ether was allowed to
diffuse slowly into a solution of analytically pure 3c in CH2Cl2.
Blue plates of 3c‚1/2H2O‚CH2Cl2 crystallized after several days at
room temperature.
Anal. (%) Calcd for C20Cl2CuH26N4O3 (504.90 g/mol): C 47.58,
H 5.19, N 11.10. Found: C 47.45, H 5.32, N 10.85. MS (FAB,
NBA): m/z ) 468 [Cu(Boc-Ala-bpa)(Cl)]+, 433 [Cu(Boc-Ala-
bpa)]+, 371 [Boc-Ala-bpa]+. IR (KBr, [cm-1]): ν ) 3436 (νNsH),
3363 (νNsH), 1690 (νCdO, amide + urethane), 1609 (νCdN). UV-
vis (CH2Cl2, λ [nm] (ꢀ)): 803 (288).
[Cu(Boc-Leu-bpa)(Cl)2] (3d). The amounts of each reagent
follow: Boc-Leu-bpa (1d) (539 mg, 1.31 mmol), CuCl2‚2H2O
(223 mg, 1.31 mmol), CH3CN (15 mL). The blue reaction mix-
ture turned green after a short period of time. The solution was
stirred overnight and the solvent removed by rotary evaporation
leaving a mixture of blue, green, and brown solids. The crude
product was dissolved in CH2Cl2 and the solution filtered to re-
move insoluble material. Slow diffusion of ether into the filtrate
for several days resulted in the crystallization of light-green plates
of 3d which were isolated by filtration and dried under vacuum
(557 mg, 1.02 mmol, 78%). Green crystalline plates of 3d‚CH2Cl2
were obtained by ether diffusion into a solution of pure 3d in CH2-
Cl2. The X-ray structure was of poor quality which was mostly
due to a severe disordering of the Boc protecting group. However,
the similarity of the first coordination sphere to that of 3a and 3c
was confirmed.
Anal. (%) Calcd for C23Cl2CuH32N4O3 (546.98 g/mol): C 50.51,
H 5.90, N 10.24. Found: C 50.02, H 5.91, N 10.09. MS (FAB,
4672 Inorganic Chemistry, Vol. 43, No. 15, 2004

Tertiary Carboxamides in Metal Complexes
3.38, N 10.78. MS (FAB, NBA): m/z ) 631 [Cu2(bpa)2Cl3]+,
297 [Cu(bpa)Cl]+, 262 [Cu(bpa)]+. IR (KBr, [cm-1]): ν ) 1609
(νCdN).
area detector (2b, 5, 5‚2H2O, 6). Absorption corrections were made
numerically from crystal faces (5‚2H2O),45 or using empirical
ψ-scans (2b),46 or with the program SORTAV (5, 6),47 or
SCALEPACK (3c).48 All structures were solved by direct methods
and refined using full-matrix least-squares on F2 using the program
packages SHELXTL NT 5.10 (2b, 5, 5‚2H2O, 6)49 and SHELXS-
97/SHELXL-97 (3c).50 Non-H-atoms were anisotropically refined.
H-atoms were localized in the density map and isotropically refined,
or fixed in geometrically calculated positions (riding mode). Details
have been deposited at the Cambridge Crystallographic Data Centre
as supplementary publications CCDC nos. 182769 (2b), 184027
(3c), 231941 (5), 231942 (5‚2H2O), and 231943 (6). Copies are
available free of charge from CCDC, 12 Union Road, Cam-
bridge CB2 1 EZ (U.K.) [Fax: (+44)1223-336033. E-mail:
deposit@ccdc.cam.ac.uk.]
Theoretical Calculations. All calculations used the Gaussian
9839 suite of programs. Geometries were fully optimized without
constraints using the Becke 3-parameter hybrid functional40 in
conjunction with the Lee-Yang-Parr correlation functional41
(B3LYP).42 The Schaefer-Horn-Ahlrichs split-valence basis set43
was used with an additional set of polarization functions44 for all
calculations. Stationary points were characterized by calculation
of their normal vibrations within the harmonic approximation. All
structures reported here are minima.
X-ray Crystallography. A summary of the crystallographic
parameters for all compounds is given in Table 3. Intensity data
(Mo KR, λ ) 0.71073 Å) were collected on a Siemens P4
diffractometer (3c; ω-scans, 6°/min) and on a Nonius Kappa CCD
Supporting Information Available: Crystallograhic data in CIF
format. This material is available free of charge via the Internet at
http://pubs.acs.org.
(39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.;
Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98; Gaussian, Inc.: Pittsburgh, PA, 1998.
(40) Becke, A. D. Phys. ReV. A 1988, 38, 3098.
IC0496774
(44) Huzinaga, S. Gaussian Basis Sets for Molecular Calculations;
Elsevier: Amsterdam, 1984. Raghavachari, K.; Trucks, G. W. J. Chem.
Phys. 1989, 91, 1062.
(45) Coppens, P. In Crystallographic Computing; Ahmed, F. R., Ed.; S.
R. Hall & C. P. Huber: Copenhagen, 1970; pp 255-270.
(46) North, A. C. T.; Phillipps, D. C. Acta Crystallogr. 1968, A24, 351.
(47) Blessing, R. H. Acta Crystallogr. 1995, A51, 33.
(48) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
(49) SHELXTL NT 5.10; Bruker AXS: Madison, WI, 1998.
(50) Sheldrick, G. M. SHELXS-97 and SHELXL-97; Universita¨t Go¨ttin-
gen: Go¨ttingen, Germany, 1997.
(41) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(42) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(43) Schaefer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571-
2577.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4673