Synthesis and characterization of cross-bridged cyclams and pendant-armed derivatives and structural studies of their copper(II) complexes

Article abstract of DOI:10.1021/ja001295j

A series of 1,4,8,11-tetraazabicyclo[6.6.2]hexadecane ligands, including the parent compound, N,N′-dialkyl variants, and the first N,N′-di-pendant-arm derivatives have been synthesized by a short, efficient, and conceptually novel approach. Their copper(I

Full text of DOI:10.1021/ja001295j

J. Am. Chem. Soc. 2000, 122, 10561-10572
10561
Synthesis and Characterization of Cross-Bridged Cyclams and
Pendant-Armed Derivatives and Structural Studies of Their Copper(II)
Complexes
Edward H. Wong,*,† Gary R. Weisman,*,† Daniel C. Hill,†,| David P. Reed,†,
Mark E. Rogers,† Jeffrey S. Condon,† Maureen A. Fagan,†,] Joseph C. Calabrese,‡
Kin-Chung Lam,§ Ilia A. Guzei,§,# and Arnold L. Rheingold§
Contribution from the Department of Chemistry, UniVersity of New Hampshire,
Durham, New Hampshire 03824-3598, the Central Research and DeVelopment Department,
E. I. DuPont Nemours & Co., Inc., Experimental Station, Wilmington, Delaware 19880-0228, and
the Crystallography Laboratory, Department of Chemistry and Biochemistry, UniVersity of Delaware,
Newark, Delaware 19716
ReceiVed April 13, 2000
Abstract: A series of 1,4,8,11-tetraazabicyclo[6.6.2]hexadecane ligands, including the parent compound, N,N′-
dialkyl variants, and the first N,N′-di-pendant-arm derivatives have been synthesized by a short, efficient, and
conceptually novel approach. Their copper(II) complexes have been prepared, and four of these were structurally
characterized by X-ray diffraction. In all four complexes, the cross-bridged tetraamine ligand was found to be
cis-folded, coordinating the metal cation within its molecular cleft using all four nitrogen lone pairs. Geometries
intermediate between idealized square pyramidal and trigonal bipyramidal coordination were found for three
of the complexes, whereas a distorted octahedral copper coordination was found for the complex of a di-
pendant-arm cross-bridged cyclam.
Introduction
members of a novel class of ligands having nonadjacent nitro-
gens bridged by (CH2)2, the “cross-bridged” tetraamines.6,7 We
designed these bicyclo[6.6.2], [6.5.2], and [5.5.2] tetraamines
to be capable of adopting conformations8,9 having all four nitro-
gen lone pairs convergent upon a cleft (in,in at the bridgehead
nitrogens10) for complexation of small metal ions and encap-
The design, synthesis, and coordination chemistry of new
polyamine ligands and pendant-arm derivatives continues to
stimulate intense research effort due to realized and potential
applications in radiopharmaceutical, catalysis, and biomimetic
chemistry.1-3 Wainwright and Hancock have investigated
“structurally-reinforced” macrocyclic tetraamines whereby ad-
jacent nitrogens (i.e., those without intervening nitrogens) are
linked by ethylene bridges to favor trans coordination of metal
cations.4,5 We have communicated the syntheses of several
(2) (a) Parker, D., Ed. Macrocyclic Synthesis. A Practical Approach;
Oxford University Press: Oxford, 1996. (b) Breitenbach, J.; Boosfeld, J.;
Vo¨gtle, F. Compr. Supramol. Chem. 1996, 2, 29. (c) Bradshaw, J. S.;
Krakowiak, K. E.; Izatt, R. M. Aza-Crown Macrocycles; In The Chemistry
of Heterocyclic Compounds; Taylor, E. C., Ed.; John Wiley and Sons: New
York, 1993: vol. 51. (d) Martell, A. E.; Hancock, R. D. Coord. Chem.
ReV. 1994, 133, 39. (e) Dietrich, B.; Viout, P.; Lehn, J.-M. Macrocyclic
Chemistry; VCH: Weinheim, 1993. (f) Busch, D. H. Chem. ReV. 1993, 93,
847. (g) Bencini, A.; Bianchi, A.; Paoletti, P.; Paoli, P. Pure Appl. Chem.
1993, 65, 381. (h) Kaden, T. A. In Crown Compounds; Cooper, S. R., Ed.;
VCH: New York, 1992; p 135. (i) Bianchi, A.; Micheloni, M.; Paoletti, P.
Coord. Chem. ReV. 1991, 110, 17. (j) Hancock, R. D.; Martell, A. E. Chem.
ReV. 1989, 89, 1875. (k) Lindoy, L. F. The Chemistry of Macrocyclic Ligand
Complexes; Cambridge University Press: Cambridge, 1989.
(3) (a) Wainwright, K. P. Coord. Chem. ReV. 1997, 166, 35. (b) Hancock,
R. D.; Maumela, H.; de Sousa, A. S. Coord. Chem. ReV. 1996, 148, 315.
(c) Bernhardt, P. V. Coord. Chem. ReV. 1990, 104, 297. (d) Kaden, Th. A.
AdV. Supramol. Chem. 1993, 3, 65.
(4) (a) Wainwright, K. P. Inorg. Chem. 1980, 19, 1396-1398. (b)
Ramasubba, A.; Wainwright, K. P. J. Chem. Soc., Chem. Commun. 1982,
277.
(5) (a) Hancock, R. D.; Gary, P.; Wade, P. W.; Hosken, G. D. Pure
Appl. Chem. 1993, 65, 473-476. (b) Hancock, R. D. In Crown Compounds;
Cooper, S. R., Ed.; VCH: New York, 1992; pp 167-190. (c) Kowallick,
R.; Neuburger, M.; Zehnder, M.; Kaden, T. A. HelV. Chim. Acta 1997, 80,
948.
* To whom correspondence should be addressed. E-mail: ehw@hypatia.
unh.edu; gary.weisman@unh.edu.
† University of New Hampshire.
‡ DuPont Nemours & Co.
§ University of Delaware.
| Present address: Astra-Zeneca Pharmaceutical, 1800 Concord Pike,
Wilmington, DE 19850-5437.
] Present address: Rhode Island College, Department of Physical
Science, 600 Mt. Pleasant Avenue, Providence, RI 02908-1991.
Present address: Chemical Development, Albany Molecular Research
Inc., 21 Corporate Circle, Albany, NY 12203.
# Present address: Department of Chemistry, Iowa State University,
Ames, IA 50010.
(1) (a) Orvig, C.; Abrams, M. J., Eds. Medicinal Inorganic Chemistry
(Thematic Issue); Chem. ReV. 1999, 99(9), 2210-2842. (b) Reichert, D.
E.; Lewis, J. S.; Anderson, C. J. Coord. Chem. ReV. 1999, 184, 3. (c)
Comblin, V.; Gilsoul, D.; Hermann, M.; Humblet, V.; Jacques, V.; Mesbahi,
M.; Sauvage, C.; Desreux, J. F. Coord. Chem. ReV. 1999, 185-186, 451.
(d) Yam, V. W.-W.; Lo, K. K.-W. Coord. Chem. ReV. 1999, 184, 157. (e)
Motekaitis, R. J.; Rogers, B. E.; Reichert, D. E.; Martell, A. E.; Welch, M.
J. Inorg. Chem. 1996, 35, 3821. (f) DeNardo, G. L.; DeNardo, S. J. In
Cancer Therapy with Radiolabeled Antibodies; Goldenberg, D. M., Ed.;
CRC Press: Boca Raton, FL, 1995; p 217. (g) Togni, A.; Venanzi, L. M.
Angew. Chem., Int. Ed. Engl. 1994, 33, 497. (h) Jurisson, S.; Berning, D.;
Jia, W.; Ma, D. Chem. ReV. 1993, 93, 1137. (i) Kimura, E. Pure Appl.
Chem. 1993, 65, 355. (j) Parker, D. In Crown Compounds; Cooper, S. R.,
Ed.; VCH Publishers: New York, 1992; p 51. (k) Kimura, E. Tetrahedron
1992, 48, 6175. (l) Parker, D. Chem. Soc. ReV. 1990, 19, 271.
(6) Weisman, G. R.; Rogers, M. E.; Wong, E. H.; Jasinski, J. P.; Paight,
E. S. J. Am. Chem. Soc. 1990, 112, 8604.
(7) Weisman, G. R.; Wong, E. H.; Hill, D. C.; Rogers, M. E.; Reed, D.
P.; Calabrese, J. C. J. Chem. Soc., Chem. Commun. 1996, 947.
(8) For discussions of flexible but conformationally well-defined mol-
ecules, see: (a) Hoffmann R. W.; Stahl, M.; Schopfer, U.; Frenking, G.
Chem. Eur. J. 1998, 4, 559. (b) Hoffmann, R. W. Angew. Chem., Int. Ed.
Engl. 1992, 31, 1123. (c) Sanders, J. K. M.Chem. Eur. J. 1998, 4, 1378.
10.1021/ja001295j CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/18/2000

10562 J. Am. Chem. Soc., Vol. 122, No. 43, 2000
Wong et al.
tetraamines having longer bridging groups19 and their Cu(II)
and Ni(II) complexes.20
Herein we present detailed procedures for the efficient
synthesis of cross-bridged cyclam ligands, including the first
pendant-armed derivatives. The rationale of the synthetic
strategy is described and structures of two alkylated tetracyclic
bisaminal synthetic intermediates are presented. Preparation and
structural characterizations of four Cu(II) complexes of the
ligands are detailed.
Figure 1. Metal complexation by cross-bridged cyclams.
sulation of protons. Such ligands may be thought of as
“clamshells” of facially coordinating triazacycloalkanes.11 The
constitutional design precludes trans coordination and ensures
that these flexible ligands may adopt low-strain medium-ring
conformations ideal for cis-folded coordination. The viability
of this concept was verified with the demonstration6 that dialkyl
cross-bridged tetraamines are proton sponges12 and Li+-selective
alkali metal ion complexers. Our communication on Cu(II) and
Ni(II) complexes appeared in 1996.7 The structural results
showed that the [6.6.2] cross-bridged cyclams indeed form cis-
folded complexes, the tetraamines adopting C2 conformations
with each of the 10-membered rings in the distorted diamond-
lattice [2323] conformation13 that was anticipated on the basis
of our ligand design (Figure 1).14 More recently, we reported
related Zn(II) complexes of cross-bridged cyclam.15 Since 1998,
Busch, Alcock and co-workers have also reported their work
on 3d-transition metal complexations16,17 using our cross-bridged
ligands and a C-hexamethyl analogue.17 Busch et al. have
described the cross-bridged ligands as “ultra-rigid”.16a,d How-
ever, because bridgehead bicyclic tetraamines having two
medium rings are conformationally flexible and inhomogeneous
(EFF Monte Carlo searches of conformational space find scores
of energetically accessible conformations18), we prefer not to
use that term. Micheloni and Ciampolini, Springborg, and others
have also reported a number of interesting cross-bridged
Results and Discussion
Ligand Synthesis and Rationale. Our short, efficient route
to the cross-bridged cyclams and selected pendant-arm deriva-
tives is shown in Scheme 1. In summary, the approach involves
(a) condensation of glyoxal and cyclam (1) to give tetracyclic
bisaminal 2,21 (b) highly regioselective dialkylation of 2 to give
bis-quaternary ammonium halides 3, and (c) reduction to give
double-ring-expanded cross-bridged cyclams 4. N,N′-Dibenzyl
cross-bridged cyclam 4a is cleanly debenzylated by atmospheric
hydrogenolysis to give the parent cross-bridged ligand 5, which
can be subsequently elaborated to pendant-arm derivatives. This
synthetic strategy avoids the problems inherent in synthesis of
10-membered rings by cyclization methodologies.2e,22
Bisaminal Alkylation. Our synthetic design and the ultimate
success of the implemented synthesis of the cross-bridged
cyclams was predicated upon the concept that the regioselec-
tivity of the dialkylation should be a consequence of the
conformation of tetracyclic bisaminal 2. As shown in Figure 2,
the global minimum of cis-fused 2 is an all-chair C2 conforma-
tion possessing a cleft.21,23 Two homotopic exo nitrogens have
lone pairs which protrude from the convex face of the molecule,
whereas two homotopic endo nitrogens have lone pairs which
are more sterically concealed on the concave face. Enantiomer-
ization of 2, which exchanges the exo and endo nitrogens, is
fast on the laboratory time scale. The barrier of this enanti-
omerization, a multistep conformational process which involves
net inversion of all four nitrogens and both piperazine rings, is
approximately 15 kcal/mol.21,24 Initial alkylation takes place on
a more sterically accessible exo nitrogen of 2 (alkylation of only
one of the enantiomers of 2 is shown in Figure 2 for simplicity).
The endo-alkylation is sterically less favorable than exo-
alkylation because the transition state of the former involves
two incipient g+g- interactions (R-N+-C-N-C and R-N+-
(9) (a) Alder, R. W. Tetrahedron 1990, 46, 683-713. (b) Alder, R. W.
Acc. Chem. Res. 1983, 16, 321-327. (c) Alder, R. W.; White, J. M. In
Conformational Analysis of Medium-Sized Ring Heterocycles; Glass, R. S.,
Ed.; VCH: New York, 1988. (d) Bernal, I., Ed. Stereochemical and
Stereophysical BehaViour of Macrocycles; Elsevier: Amsterdam, 1987.
(10) Alder, R. W.; East, S. P. Chem. ReV. 1996, 96, 2097.
(11) (a) Bhula, R.; Osvath, P.; Weatherburn, D. C. Coord. Chem. ReV.
1988, 91, 89. (b) Chaudhuri, P.; Wieghardt, K. Prog. Inorg. Chem. 1987,
35, 329. (c) Wieghardt, K. Pure Appl. Chem. 1988, 60, 509.
(12) (a) Alder, R. W. Chem. ReV. 1989, 89, 1215-23. (b) Staab, H. A.;
Saupe, T. Angew. Chem., Int. Ed. Engl. 1988, 27, 865. (c) Bencini, A.;
Bianchi, A.; Garcia-Espan, E.; Micheloni, M.; Ramirez, J. A. Coord. Chem.
ReV. 1999, 188, 97.
(13) Dale, J. Acta Chem. Scand. 1973, 27, 1115. Dale, J. In Topics in
Stereochemistry; Allinger N. L., Eliel, E. L., Eds.; Wiley: New York, 1976;
vol 9, p 199. Dale’s nomenclature utilizes [abc...] to represent ring
conformation, where a, b, and c are each integers denoting the number of
bonds between two true corners. A sequence of two gauche (synclinal)
torsion angles of the same sign introduces a bend; the atom flanked by
these torsion angles is referred to as a true corner. The global minimum
conformation of cyclodecane has four true corners and four sides; it is a
“rectangular” diamond-lattice [2323] conformation.
(14) A [5.5.2] cross-bridged cyclen was reported by others in 1994, but
the only complex had externally bound Cu(II): Bencini, A.; Bianchi, A.;
Bazzicalupi, C.; Ciampolini, M.; Fusi, V.; Micheloni, M.; Nardi, N.; Paoli,
P.; Valtancoli, B. Supramolecular Chem. 1994, 3, 141.
(19) (a) Bencini, A.; Fusi, V.; Giorgi, C.; Micheloni, M.; Nardi, N.;
Valtancoli, B. J. Chem. Soc., Perkin Trans. 2 1996, 2297. (b) Bencini, A.;
Bianchi, A.; Bazzicalupi, C.; Ciampolini, M.; Dapporto, P.; Fusi, V.;
Micheloni, M.; Nardi, N.; Paoli, P.; Valtancoli, B. J. Chem. Soc., Perkin
Trans. 2 1993, 115. (c) Bencini, A.; Bianchi, A.; Bazzicalupi, C.; Ciampolini,
M.; Dapporto, P.; Fusi, V.; Micheloni, M.; Nardi, N.; Paoli, P.; Valtancoli,
B. J. Chem. Soc., Perkin Trans. 2 1993, 715. (d) Springborg, J.; Pretzmann,
U.; Nielsen, B.; Olsen, C. E.; Søtofte, I. Acta Chem. Scand. 1998, 52, 212.
(e) Springborg, J.; Kofod, P.; Olsen, C. E.; Toftlund, H.; Søtofte, I. Acta
Chem. Scand. 1995, 49, 547. (f) Denat, F.; Lacour, S.; Brande`s, S.; Guilard,
R. Tetrahedron Lett. 1997, 38, 4417. (g) Brande`s, S.; Lacour, S.; Denat,
F.; Pullumbi, P.; Guilard, R. J. Chem. Soc., Perkin Trans. 1 1998, 639.
(20) (a) Springborg, J.; Sotofte, I. Acta Chem. Scand. 1997, 51, 357. (b)
Springborg, J.; Glerup, J.; Sotofte, I. Acta Chem. Scand. 1997, 51, 832. (c)
Bencini, A.; Bianchi, A.; Borselli, A.; Ciampolini, M.; Micheloni, M.; Paoli,
P.; Valtancoli, B.; Garcia-Espana, E.; Ramirez, J. A. J. Chem. Soc., Perkin
Trans. 2 1990, 209. (d) Sanzenbacher, R.; Sotofte, I.; Springborg, J. Acta
Chem. Scand. 1999, 53, 457.
(15) Niu, W.; Wong, E. H.; Weisman, G. R.; Lam, K.-C.; Rheingold,
A. L. Inorg. Chem. Commun. 1999, 2, 361.
(16) (a) Hubin, T. J.; McCormick, J. M.; Collinson, S. R.; Alcock, N.
W.; Busch, D. H. J. Chem Soc., Chem. Commun. 1998, 1675. (b) Hubin,
T. J.; Alcock, N. W.; Busch, D. H. Acta Crystallogr. 1999, C55, 1404. (c)
Hubin, T. J.; Alcock, N. W.; Busch, D. H. Acta Crystallogr. 2000, C56,
37. (d) Hubin, T. J.; McCormick, J. M.; Collinson, S. R.; Buchalova, M.;
Perkins, C. M.; Alcock, N. W.; Kahol, P. K.; Raghunathan, A.; Busch, D.
H. J. Am. Chem. Soc. 2000, 122, 2512.
(21) Weisman, G. R.; Ho, S. C.-H.; Johnson, V. B. Tetrahedron Lett.
1980, 21, 335.
(22) Mandolini, L. AdV. Phys. Org. Chem. 1986, 22, 1.
(23) Gluzinski, P.; Krajewski, J. W.; Urbanczyk-Lipkowska, Z.; Bleidelis,
J.; Kemme, A. Acta Crystallogr. 1982, B38, 3038.
(17) Hubin, T. J.; McCormick, J. M.; Alcock, N. W.; Clase, H. J.; Busch,
D. H. Inorg. Chem. 1999, 38, 4435.
(18) Weisman, G. R.; Hill, D. C.; Reed, D. P.; Hines, M. S.; Rogers, M.
E.; Wong, E. H.; Jasinski, J. P.; Paight, E. S.; Lam, K.-C.; Rheingold, A.
L. Manuscript in preparation.
(24) (a) Kolinski, R. A.; Riddell, F. G. Tetrahedron Lett. 1981, 22, 2217.
(b) Riddell, F. G.; Murray-Rust, P.; Kolinski, R. A.; Gluzinski, P.
Tetrahedron 1982, 38, 673.

Cross-Bridged Cyclams and Pendant-Armed DeriVatiVes
J. Am. Chem. Soc., Vol. 122, No. 43, 2000 10563
Scheme 1
Figure 3. X-ray structure of 3a.
In practice, the dialkylations to give 3 were very clean reactions
under our experimental conditions (MeCN, room temperature);
there was no evidence for the presence of other constitutional
isomers in the crude reaction product. (However, it should be
noted that increasing the temperature resulted in impure products
and lower yields; vide infra.) NMR results for 3a and 3b are
completely consistent with the C2 exo,exo structure. The X-ray
crystal structure of 3a (Figure 3) confirms this.27An X-ray
structure of 3b has also been reported recently.17
Implicit in the above discussion is the assumption that the
initial stereoselective alkylation of 2 is kinetically controlled
rather than thermodynamically controlled. However, the question
of kinetic vs thermodynamic control must be considered if alkyl
halides (rather than tosylates, triflates, etc.) are employed for
the alkylation of tertiary amines; halides are good nucleophiles,
making such SN2 alkylations reversible under certain conditions.
Although we expected that the reactions reported herein would
be kinetically controlled under the conditions of our experiments,
it remained to be demonstrated. Monobenzyl salt 9a was cleanly
generated by alkylation of 2 with benzyl bromide in toluene,
dibenzylation being prevented by precipitation of 9a from
Figure 2. Alkylation of tetracyclic bisaminal 2.
C-C-N). The exo-alkylated product (9) is conformationally
locked by the quaternary nitrogen, which precludes any con-
formational changes other than to high-energy twist-boat
conformations. The highly regioselective second alkylation,
which is qualitatively slower than the first, occurs on the
remaining exo nitrogen, which is less sterically hindered than
the two remaining endo nitrogens. exo-Alkylation also mini-
mizes Coulombic repulsion relative to alkylation at either of
the endo nitrogens of 9. The relative importance of these factors
in determining the regioselectivity is unknown. The high
stereoselectivity of the first step suggests that the steric factor
alone could account for the results of the second alkylation,
but it is also known that dialkylation of monocyclic aminals
(e.g. dimethylhexahydropyrimidine) requires extreme condi-
tions,25,26 so clearly, both factors favor the desired product 3.
(26) (a) Bo¨hme, H.; Da¨hne, M. Liebigs Ann. Chem. 1969, 723, 41. (b)
Bo¨hme, H.; Osmers, K. Chem. Ber. 1972, 105, 2237. (c) Carpenter, A. J.;
Chadwick, D. J. Tetrahedron 1985, 41, 3803.
(25) Duhamel, L. In Supplement F: The Chemistry of Amino, Nitroso,
and Nitro Compounds and their DeriVatiVes; Patai, S., Ed. John Wiley and
Sons: New York, 1982; Part 2, Chapter 20, p 849.
(27) During the preparation of this paper and after the completion of
our work, a structure of the dihydrate of 3a appeared.28

10564 J. Am. Chem. Soc., Vol. 122, No. 43, 2000
Wong et al.
endo nitrogen of the unalkylated aminal functionality of 9b are
more basic than the respective exo nitrogens, as reported by
Busch and Alcock.17 (It should be noted that endo-protonation
is not sterically demanding and is under thermodynamic control.)
The bond lengths of di-endo-protonated 217 exhibit the expected
structural anomeric effects (C-N shortened and C-N+ length-
ened).31 An attempt to equilibrate 9a with 11a (∼60 °C, 16 h)
led only to decomposition of 9a, which is consistent with our
observations that the dialkylations of 2 are cleaner at room tem-
perature than at higher temperatures. We believe this decom-
position likely involves proton-transfer disproportionation through
an iminium ion (vide infra) to give 9a‚HBr and a triaminoet-
hylene, but we cannot rule out the possibility that the initial
step of the decomposition is equilibration to 11a. We expect
11a to be more labile than 9a, because it is stereoelectronically
set up for opening to an iminium ion (ap lp-N-C-N+).
Interestingly, Busch and Alcock have reported that reaction
of methyl iodide with hindered hexamethyl-bisaminal 12 under
our conditions (rt) gave endo-methylated product; use of forcing
conditions (100 °C, 24 h) gave endo-methylated, exo-protonated
diiodide 13 (major) and dimethylated diiodide (minor). This may
be a consequence of steric inhibition of exo-methylation by
flanking C-methyls, that is, a kinetic effect. It must be noted,
however, that the sample used for the X-ray structure of 13
was derived from the high-temperature methylation, where
thermodynamic control is a possibility. The fact that hydroiodide
13 was obtained provides further support for the hypothesis that
monoalkylated 2 decomposes via proton-transfer disproportion-
ation.
Figure 4. X-ray structure of 9a.
solution (kinetic control).29,30 The exo-benzylated structure of
9a was verified by X-ray crystallography (Figure 4). When a
solution of 9a in MeCN was treated with excess p-methylbenzyl
bromide under our standard conditions for conversion of 2 to
3a (MeCN, room temperature, typical concentrations) but for a
period of only 1 h, 1H and 13C NMR analysis (after the removal
of the excess alkylating agent by washing) showed the crude
product to be a mixture of two compounds: the starting 9a and
unsymmetrical dialkylated salt 10. The fact that no mono- or
di-p-methylbenzylated products were observed demonstrates that
the second alkylation of 9a is much faster than debenzylation,
proving that the first benzylation is, indeed, under strict kinetic
control.
To shed some additional light on the question of the relative
thermodynamic stabilities of exo- and endo-alkylated 2 (9 and
11 respectively), as well as 14 and 15, we have carried out a
series of molecular mechanics and density functional theory
(DFT) calculations. As might be expected, a variety of popular
force fields place 9a/b well below 11a/b, because the force fields
are generally well-parametrized for steric interactions but are
not properly parametrized for the R-amino ammonium ion
(alkylated aminal) functionality. Of the force fields that we
investigated, the Merck Molecular Force Field (MMFF)33 gave
the smallest 11-9 (endo-exo) energetic differences, placing
the 9a cation (exo) 2.9 kcal/mol below the 11a cation (endo)
and the 9b cation (exo) 2.0 kcal/mol below the 11b cation
(endo). It also places exo-methyl cation 15 6.1 kcal/mol aboVe
endo-methyl analogue 14. MMFF also gave geometries that
were in reasonable accord with the available crystallographic
results. Density functional calculations were carried out by the
Although neither we6,7 nor others17,28,30 have observed any
endo-alkylation products of 2 (indeed, we had hoped not to!),
the question of the relative thermodynamic stabilities of exo
salts 9 and their endo analogues 11 is certainly very interesting.
Although the former alkylated aminals are favored on steric
grounds, the latter are favored stereoelectronically. In 11, the
remaining nitrogen lone pair of the alkylated aminal unit is
antiperiplanar (ap) to the aminal C-N+ bond, the arrangement
that maximizes the stabilizing hyperconjugative nNfσ* interac-
tion. The corresponding lp-N-C-N+ torsion angle in 9 is
synclinal (sc), a less stabilizing geometry. Alder and one of us
have documented studies of such stereoelectronic effects in
aminals, protonated aminals, and alkylated aminals.31,32 This
type of stabilization (a generalized anomeric effect) is no doubt
responsible for the fact that the endo nitrogens of 2 and the
(31) (a) Alder, R. W.; Carniero, T. M. G.; Mowlam, R. W.; Orpen, A.
G.; Petillo, P. A.; Vachon, D. J.; Weisman, G. R.; White, J. M. J. Chem.
Soc., Perkin Trans. 2 1999, 589. (b) Alder, R. W. Tetrahedron 1990, 46,
683.
(28) Kotek, J.; Hermann, P.; VojtÄisek, P.; Rohovec, J.; Lukes, I. Collect.
Czech. Chem. Commun. 2000, 243.
(29) Hill, D. C., Ph.D. Dissertation, University of New Hampshire, 1995;
Diss. Abstr. Int., B 1996, 57(2), 1083.
(32) Alder, R. W.; Heilbronner, E.; Honegger, E.; McEwen, A. B.; Moss,
R. E.; Olefirowicz, E.; Petillo, P. A.; Sessions, R. B.; Weisman, G. R.;
White, J. M.; Yang, Z. J. Am. Chem. Soc. 1993, 115, 6580.
(33) Halgren, T. A. J. Comput. Chem. 1996, 17, 490 and following papers
in issue.
(30) Kolinski, R. A. Pol. J. Chem. 1995, 69, 1039.

Cross-Bridged Cyclams and Pendant-Armed DeriVatiVes
J. Am. Chem. Soc., Vol. 122, No. 43, 2000 10565
the reductions of 3 are slow under these conditions, they are
remarkably clean. The use of higher temperatures or larger
proportions of water generally led to impure product, presumably
as a result of iminium hydrolysis. Salts 3 are not reduced at an
appreciable rate by LiAlH4/Et2O, which suggests that the more
polar medium is necessary for intermediate iminium ion
formation. It should be noted that unalkylated 2 is resistant to
NaBH4 or LiAlH4 reduction and is reduced to 1,5,8,12-
tetraazabicyclo[10.2.2]hexadecane by DIBALH.39 The reduction
mechanism proposed in Figure 5 is supported by the fact that
NaBD4 reduction is stereoselective but not stereospecific.40 We
suggest that the rate-determining step of the reaction is the first
step, in which tetracyclic 3 is converted to a tricyclic iminium
ion via a twist-boat conformation of 3 in order to achieve the
ap lp-N-C-N+ torsion angle that is stereoelectronically ideal
for cleavage.
Figure 5. Mechanism for double-reductive ring expansion of 3.
perturbative Becke-Perdew34 method implemented in Spartan
V5.1.3,35 using the DN** numerical basis set, which is similar
to 6-31G**. pBP/DN**//MMFF calculations place the 9a cation
(exo) 1.3 kcal/mol below the 11a cation (endo), but the 9b cation
(exo) 0.2 kcal/mol above the 11b cation (endo). pBP/DN**//
MMFF on cations 14 and 15 gave the same result as the MMFF
calculations: endo 14 is 6.1 kcal/mol more stable than 15.
Geometry optimizations of the 9b cation (exo) and the 11b
cation (endo) at the DFT level (pBP/DN**//pBP/DN**) were
also carried out, yielding the result that the latter (endo) is more
stable by 2.1 kcal/mol. However, the C(aminal)-N+ bond length
in the geometry-optimized 11 cation was unusually long (1.66
Å), and it has been pointed out that DFT calculations tend to
overestimate anomeric effects,36 so these computational results
must be interpreted with appropriate caution. Nevertheless, the
calculations are in qualitative accord with the endo-methylation
of 12 observed by Busch and Alcock (whether it is a kinetic or
a thermodynamic process). The calculations also suggest to us
that endo-benzylation of 2 is unlikely under any conditions but
that thermodynamic endo-methylation of 2 cannot be excluded
as a possibility.
4a and 4b are remarkably basic (proton sponges).6,7 Although
this aspect of the behavior of the cross-bridged tetraamines will
be detailed in the future,18 it is worth pointing out here that
dissolution of N,N′-dialkyl cross-bridged cyclams in chloroform
for other than brief periods (even for NMR spectroscopy) should
be avoided, because they react.
Debenzylation and N,N′-Pendant-Arming. Debenzylation
of 4a was achieved by catalytic hydrogenolysis (1 atm) using
10% Pd/C in glacial acetic acid. The crude product after workup
(benzene extraction) was >99% pure by NMR in some reactions
but may be recrystallized from Et2O at low temperature (-78
°C) if necessary. Compound 5 is a crystalline, sublimable solid
that is soluble not only in water and polar aprotic solvents
(DMSO, CH3CN) but also in nonpolar solvents, including
pentane and CCl4. This suggests that the N-H hydrogens form
N-H‚‚‚N transannular hydrogen bonds to the bridgehead tertiary
amino nitrogens, thus allowing the molecule to present a
lipophilic exterior to solvent. The 1H NMR spectra of 5 in C6D6,
CDCl3, and CD3CN (δN-H ) 3.5-3.7) and IR spectra (N-H
stretch of 3260 cm-1 invariant over concentration range of
0.016-0.13 M 5 in CHCl3) are consistent with intramolecular
H-bonding. Compound 5 is also considerably less basic than
either 4a or 4b.18
Parent cross-bridged cyclam 5 has been converted to a number
of N,N′-difunctionalized pendant-arm derivatives by standard
secondary amine functionalization routes. The derivatives that
were used as ligands in this study are included in Scheme 1.
One unanticipated result was encountered: an attempt to convert
diester 6 to diamide 7 by reaction with aqueous ammonia under
standard conditions gave diacid 8 (after pH adjustment) instead.
We surmise that the basicity of the cross-bridged moiety is the
cause of the unexpected reaction, generating sufficient hydroxide
to compete nucleophilically with the ammonia.
Preparation and Spectral Characterization of Copper(II)
Complexes. Copper(II) complexes of these cross-bridged ligands
were readily prepared in methanol or ethanol solvents. These
include Cu(ClO4)2‚5, CuCl2‚5, Cu(ClO4)2‚4b, CuCl2‚4a, Cu‚8,
and Cu(ClO4)2‚7. Elemental analyses were consistent with the
formation of 1:1 complexes in each case. In addition, the
following spectral data were obtained.
Reductive Ring Expansion. The purpose of the regioselec-
tive dialkylation that was discussed in the previous section is
to set the system up for double reductive ring expansion via
iminium ions to the bicyclo[6.6.2] ring system (Figure 5). Ring
expansion approaches to medium rings and polycyclic systems
have been reviewed.37 Prior to our work, Alder and co-workers
had ingeniously used reductive cleavage of tricyclic R-amino
ammonium salts (alkylated aminals) to prepare a series of
medium-ring bridgehead bicyclic diamines.38 They generally
used LiAlH4 in ether solvents for their reductions, but we found
NaBH4 in 95% EtOH at room temperature to be the system of
choice for preparation of the cross-bridged cyclams. Although
(34) (a) Becke, A. D. Phys. ReV. A: At., Mol., Opt. Phys. 1988, 38,
3089. (b) Perdew, J. P. Phys. ReV. B: Condens. Matter Mater. Phys. 1986,
33, 8822.
(35) Spartan V5.1.3: Wavefunction, Inc., 18401 Von Karman, Suite 370,
Irvine, CA 92612.
(36) (a) St.-Amant, A.; Cornell, W. D.; Kollman, P. A.; Halgren, T. A.
J. Comput. Chem. 1995, 16, 1483. (b) Hehre, W. J.; Lou, L. A Guide to
Density Functional Calculations in Spartan; Wavefunction: Irvine, CA,
1997.
(38) (a) Alder, R. W.; Eastment, P.; Hext, N. M.; Moss, R. E.; Orpen,
A. G.; White, J. M. J. Chem. Soc., Chem. Commun. 1988, 1528. (b) Alder,
R. W.; Eastment, P.; Moss, R. E.; Sessions, R. B.; Stringfellow, M. A.
Tetrahedron Lett. 1982, 23, 4181. (c) Alder, R. W.; Sessions, R. B.
Tetrahedron Lett. 1982, 23, 1121. (d) Alder, R. W.; Sessions, R. B.;
Gmuender, J. O.; Grob, C. A. J. Chem. Soc., Perkin Trans. 2 1984, 411.
(e) Alder, R. W.; Casson, A.; Sessions, R. B. J. Am. Chem. Soc. 1979, 101,
3651.
(37) Hesse, M. Ring Enlargement in Organic Chemistry; VCH: Wein-
heim, 1991.
(39) Yamamoto, H.; Maruoka, K. J. Am. Chem. Soc. 1981, 103, 4186.
(40) Weisman, G. R.; Hines, M. S. Manuscript in preparation.

10566 J. Am. Chem. Soc., Vol. 122, No. 43, 2000
Wong et al.
Table 1. Summary of Electronic Spectral Data for the Copper
Complexes
tion environments. For comparison, Springborg and co-workers
have prepared related trimethylene cross-bridged cyclen copper-
(II) complexes that are five-coordinate in the solid-state as well
as in solution.20a,b Their aqueous solution electronic spectra have
d-d bands ranging from 595 to 633 nm, depending on the lone
ancillary ligand. The strongest ligand field for our copper
complexes was found with the parent ligand 5 (567 nm),
whereas the N,N′-dibenzyl ligand 4a has the weakest (680 nm)
ligand field, with the N,N′-dimethyl ligand 4b between (618
nm) them. A similar trend has been observed for copper(II)
complexes of cyclen and its tetra-N-substituted derivatives where
the ligand field strengths are ranked in the order Bn4-cyclen
(614 nm) < Me4-cyclen (602 nm) < cyclen (590 nm).45
EPR Data. In the room-temperature solid-state EPR spectrum
of Cu(ClO4)2‚5, a single broad signal with g ) 2.104 was
observed. In a frozen methanol matrix, however, an axial
spectrum with g| ) 2.271, and g ) 2.146 was obtained. For
the complex CuCl2‚5, its solid-state EPR spectrum at room
temperature is rhombic with g1 ) 2.171, g2 ) 2.087, and g3 )
2.050. By contrast, its frozen methanol solution spectrum is quite
complex, with the presence of ∆M ) 2 transitions indicative
of the possible existence of a dimeric species.46 At room
temperature in the solid state, an approximately axial EPR
spectrum was observed for the complex CuCl2‚4a, with g| )
2.234 and g ) 2.094. In a frozen methanol matrix at 77 K, the
signals are at g| ) 2. 284 and g ) 2.098. Only a single broad
EPR signal was observed for Cu‚8 at room temperature in a
solid sample, with g ∼ 2.06. In frozen methanol, the EPR
spectrum is near axial, with g| ) 2.289 and g ) 2.096. Detailed
simulation and assignment of these spectra have not been
completed. However, all have g| > g > 2.0, which is consistent
with Jahn-Teller distorted square-pyramidal or octahedral
geometries.47
electronic absorption maxima,
complex
λ
max (ꢀ, Μ-1 cm-1
)
Cu(ClO4)2‚5
in MeCN
solid state
CuCl2‚5
in MeCN
solid state
Cu(ClO4)2‚4b
in MeCN
solid state
CuCl2‚4a
in MeCN
solid state
Cu‚8
567 nm (120)
582 nm
596 nm (196)
579 nm
618 nm (76)
625 nm
680 nm (75)
675 nm
in MeOH
solid state
Cu(ClO4)2‚7
in MeCN
solid state
642 nm (20)
629 nm
630 nm (24)
625 nm
Infrared Spectra. In the spectrum of Cu(ClO4)2‚5, a slightly
broadened N-H stretching band at 3269 cm-1 of medium
intensity was observed (parent ligand νNH is at 3260 cm-1). The
very strong and broad perchlorate band that was centered at
approximately 1090 cm-1 was not sufficiently resolved to be
of diagnostic value for the perchlorate coordination modes found
to be present in the X-ray structure of this complex (vide infra).41
The IR spectrum of CuCl2‚5 revealed the presence of lattice
H2O by a broad, medium-intensity band at 3438 cm-1. A
broadened N-H stretch was also found at 3212 cm-1. A broad
water of hydration band of medium intensity was observed
between 3400 and 3470 cm-1 in the IR spectrum of CuCl2‚4a.
For the pendant-armed ligand complex Cu‚8‚NaClO4, no IR
absorptions attributable to COOH groups (usually between 1720
and 1740 cm-1 for the protonated ligand) were observed.
Instead, carboxylate bands at 1590, 1616 cm-1 were noted,
which is consistent with metal coordination. Known carboxym-
ethyl pendant-armed macrocyclic tetraamine complexes of
X-ray Structural Studies of the Copper(II) Complexes Cu-
(ClO4)2‚5, CuCl2‚5, CuCl2‚4a‚H2O, and [Cu‚8‚Na(H2O)-
ClO4]2‚H2O. As predicted on the basis of molecular mechanics
calculations,48 each Cu(II) complex structure has four convergent
ligand nitrogen donors coordinating the Cu(II) that is inside the
molecular cleft in a cis-V-folded coordination configuration,49
with the ligand in a slightly distorted [2323]/[2323] diamond-
lattice conformation. The crystal structures of Cu(ClO4)2 ‚5
(Figure 6), CuCl2‚5 (Figure 7), and CuCl2‚4a‚H2O (Figure 8)
all exhibit five-coordinate copper. As shown in Figure 6, only
one of the perchlorates in Cu(ClO4)2 ‚5 is directly coordinated
[Cu(1)-O(1) at 2.166(6) Å], although the second is H-bonded
to a secondary amine hydrogen, N(2)-H(2N). In addition, a
weak Cu(1)‚‚‚O(2A) interaction [2.749(6) Å] to the coordinated
perchlorate of another cationic unit blocks the sixth potential
coordination site (see Supporting Information). The axial
secondary amine N(2)-Cu(1) bond lengths of 1.961(7) Å and
N(4)-Cu(1) of 1.970(7) Å are, as expected,50 significantly
copper(II) have these bands between 1570 and 1640 cm-1 42
Typical copper glycinate νCO bands range from 1593 to 1607
cm-1 43
Broad OH bands from the waters of hydration were
.
.
also found at 3580 and 3365 cm-1. An intense, broad perchlorate
absorption at 1098 cm-1 was of little diagnostic value for the
anion’s coordination mode. In the IR spectrum of the amide-
armed ligand complex Cu(ClO4)2‚7, a coordinated amide
carbonyl band at 1665 cm-1 was observed which was red-shifted
from the free-ligand value of 1685 cm-1. Kaden has previously
reported νCO’s of 1580-1660 cm-1 for coordinated amide-
pendant arms of tetraamine macrocyclic complexes.44
Electronic Spectra. Solid-state and solution electronic spectra
of the described copper complexes are summarized in Table 1.
In the solid phase, broad ligand-field bands were observed with
maxima ranging from 579 to 675 nm, which is consistent with
the blue-blue-green colors of these complexes. Their acetoni-
trile or methanol solution d-d transitions were found between
567 and 680 nm, with each spectrum reasonably similar to the
corresponding solid-state data, which suggests similar coordina-
(45) (a) Styka, M. C.; Smierciak, R. C.; Blinn, E. L.; DeSimone, R. E.;
Passariello, J. V. Inorg. Chem., 1978, 17, 82. (b) Webb, R. L.; Mino, M.
L.; Blinn, E. L.; Pinkerton, A. A. Inorg. Chem. 1993, 32, 1396.
(46) (a) Eaton, S. S.; More, K. M.; Sawant, B. M.; Eaton, G. R. J. Am.
Chem. Soc. 1983, 105, 6560. (b) Smith, T. D., Martell, A. E., J. Am. Chem.
Soc. 1972, 94, 4123.
(41) Nakamoto, K. Infrared Spectra of Inorganic and Coordination
Compounds; Wiley: New York, 1986; p 251.
(42) (a) Belsky, V. K.; Streltsova, N. R.; Kuzmina, E. N.; Nazarenko,
A. Y. Polyhedron 1993, 12, 831. (b) Riesen, A.; Zehnder, M.; Kaden, T.
A. HelV. Chim. Acta 1986, 69, 2067, 2074.
(43) Nakamoto, K.; Murito, Y.; Martell, A. E. J. Am. Chem. Soc. 1961,
83, 4528.
(47) Pilbrow, J. R. Transition Ion Electron Paramagnetic Resonance;
Clarendon Press: Oxford, 1990; pp 624-626.
(48) (a) MMX Force Field: Gajewski, J. J.; Gilbert, K. E.; McKelvey,
J. In AdVances in Molecular Modeling; Liotta, D., Ed.; JAI Press:
Greenwich, CT, 1990; Vol. 2, p 65. (b) PCMODEL v.5: Serena Software,
Box 3076, Bloomington, IN, 47402-3076.
(49) Bosnich, B.; Poon, C. K.; Tobe, M. L. Inorg. Chem. 1965, 4, 1102.
(50) Myerstein, D. Coord. Chem. ReV. 1999, 185-186, 141.
(44) Kaden, T. A. Top. Curr. Chem. 1984, 121, 157.

Cross-Bridged Cyclams and Pendant-Armed DeriVatiVes
J. Am. Chem. Soc., Vol. 122, No. 43, 2000 10567
Table 2. Selected Bond Distances (Å) and Bond Angles (deg) for
Cu(ClO4)2‚5
Cu(1)-N(2)
Cu(1)-N(3)
Cu(1)-O(1)
1.961(7) Cu(1)-N(4)
2.081(7) Cu(1)-N(1)
2.166(6)
1.970(7)
2.138(6)
N(2)-Cu(1)-N(4) 179.7(3)
N(2)-Cu(1)-N(3)
N(2)-Cu(1)-N(1)
N(3)-Cu(1)-N(1)
N(4)-Cu(1)-O(1)
86.6(3)
92.9(3)
88.5(3)
91.8(3)
N(4)-Cu(1)-N(3)
N(4)-Cu(1)-N(1)
N(2)-Cu(1)-O(1)
93.7(3)
87.1(3)
87.9(3)
N(3)-Cu(1)-O(1) 156.7(2)
N(1)-Cu(1)-O(1) 114.5(2)
for square pyramidal (C4V)], we obtained a value of 0.38 {τ )
[N(2)-Cu(1)-N(4) angle minus O(1)-Cu(1)-N(3) angle]/60}
here.51 Alternatively, the dihedral δ angle criterion of Muetterties
and Guggenberger can be used. In this approach, the angle
between the normals to the N(2)-O(1)-N(3) and N(4)-O(1)-
N(3) faces is used to gauge the distortion along the Berry
intramolecular exchange process from idealized trigonal bipy-
ramidal (TBPY, 53.1°) to idealized square pyramidal (SP, 0°).52
The δ angle is found to be 25° for this structure, which therefore
lies almost halfway between the two idealized five-coordinate
geometries. The longer equatorial N(1)-Cu(1) bond length of
2.138(6) Å [as compared to N(3)-Cu(1) of 2.081(7) Å] then
reflects the weaker apical bonding of N(1), which can be viewed
as the apex of a distorted square pyramid, while N(2), N(3),
N(4), and O(1) make up the four basal donor set. Other relevant
bond distances and angles are listed in Table 2.
Figure 6. X-ray structure of the Cu(ClO4)2‚5 complex.
The crystal structure of the CuCl2‚5 complex is shown in
Figure 7. One of the chlorides, Cl(1), is also directly coordinated,
although the second, Cl(2), is not but is, instead, H-bonded to
associated waters of hydration. A potential sixth copper
coordination site is blocked by a water molecule, which in turn
H-bonds to both chlorides. The secondary nitrogen-Cu bond
lengths of 1.996(5) and 2.003(5) Å are quite similar, with a
nearly linear N(1)-Cu-N(8) bond angle of 176.9(2)°. Again,
both of the two tertiary N-Cu bond lengths are longer as well
as significantly different from each other, with N(4)-Cu at
2.140(5) Å and N(11)-Cu at 2.081(6) Å.
Figure 7. X-ray structure of the CuCl2‚5‚3H2O complex.
Addison and Reedjik’s τ-parameter here gives an intermediate
value of 0.44, whereas the dihedral δ angle criterion of Muetter-
ties and Guggenberger using the angle between the normals to
the N(1)-Cl(1)-N(11) and N(8)-Cl(1)-N(11) faces gives a
value of 29°,51,52 again almost halfway between the two idealized
geometries. The significantly longer N(4)-Cu(1) bond allows
N(4) to be ascribed to the less strongly coordinating apex of a
distorted square pyramid. Donor atoms N(1), N(8), N(11), and
Cl(1) then represent the basal set. A recently reported structure
of a CuCl2 complex of the related cross-bridged ligand 4b by
Busch and co-workers has a very similar coordination geometry
that was described as square pyramidal (actual τ ) 0.32).17
Comparison of our CuCl2‚5 structure to a related trimethylene
cross-bridged cyclen copper bromide complex shows that the
latter’s reported coordination geometry (δ ) 31°) is also an
intermediate one.20a A significant difference between our (CH2)2
cross-bridged complex and the (CH2)3 cross-bridged analogue
is found in the substantially smaller equatorial N-Cu-N angle
of 87.8(2)°, as compared to the latter’s 99.0(3)°. Relevant bond
distances and angles are presented in Table 3.
Figure 8. X-ray structure of the CuCl2‚4a‚H2O complex.
In the crystal structure of CuCl2‚4a‚H2O (Figure 8), the sixth
potential coordination site is blocked by what may be described
as an agostic interaction between Cu(1) and an ortho-hydrogen,
shorter than the equatorial tertiary N(1)-Cu(1) bond lengths
of 2.138(6) Å and N(3)-Cu(1) of 2.081(7) Å.
Applying Addison and Reedjik’s τ-parameter for describing
five-coordinate geometries along the Berry rearrangement
between a trigonal bipyramid and a square pyramid [for which
τ ) (â-R)/60, with τ ) 1 for trigonal bipyramidal (D3h) and 0
(51) Addison, A. W.; Rao, T. N.; Reedjik, J.; van Rijn, J.; Verschoor,
G. C. J. Chem. Soc., Dalton Trans. 1984, 1349.
(52) Muetterties, E. L.; Guggenberger, L. J. J. Am. Chem. Soc. 1974,
96, 6, 1748.

10568 J. Am. Chem. Soc., Vol. 122, No. 43, 2000
Wong et al.
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) for
CuCl2‚5
Table 5. Selected Bond Distances (Å) and Bond Angles (deg) for
[Cu‚8‚Na(H2O)ClO4]2‚H2O
Cu(1)-Cl(1)
Cu(1)-N(1)
Cu(1)-N(4)
2.304(2) Cu(1)-N(8)
2.003(5) Cu(1)-N(11)
2.140(5)
1.996(5)
2.081(6)
Cu(1)-O(1)
Cu(1)-O(22)
Cu(2)-O(1′)
Cu(2)-O(22′)
Cu(1)-N(4)
Cu(1)-N(8)
2.301(2) Cu(1)-N(11)
2.007(2) Cu(1)-N(14)
2.327(2) Cu(2)-N(4′)
1.998(2) Cu(2)-N(8′)
2.034(2) Cu(2)-N(11′)
2.224(2) Cu(2)-N(14′)
2.027(2)
2.073(2)
2.050(2)
2.245(2)
2.043(2)
2.068(3)
Cl(1)-Cu(1)-N(1)
Cl(1)-Cu(1)-N(4) 122.0(1)
Cl(1)-Cu(1)-N(8)
Cl(1)-Cu(1)-N(11) 150.2(2)
N(1)-Cu(1)-N(4) 85.0(2)
90.3(2)
N(1)-Cu(1)-N(8) 176.9(2)
N(1)-Cu(1)-N(11) 92.3(2)
N(4)-Cu(1)-N(8) 93.0(2)
N(4)-Cu(1)-N(11) 87.8(2)
N(8)-Cu(1)-N(11) 85.2(2)
92.8(2)
O(1)-Cu(1)-O(22)
O(1′)-Cu(2)-O(22′) 91.70(8) O(1′)-Cu(2)-N(8′)
93.50(8) O(1′)-Cu(2)-N(4′)
80.28(9)
169.43(9)
O(1)-Cu(1)-N(4)
O(1)-Cu(1)-N(8)
O(1)-Cu(1)-N(11)
O(1)-Cu(1)-N(14)
O(22)-Cu(1)-N(4)
O(22)-Cu(1)-N(8)
81.33(9) O(1′)-Cu(2)-N(11′) 102.78(9)
171.09(8) O(1′)-Cu(2)-N(14′)
99.02(9) O(22′)-Cu(2)-N(4′)
88.13(9) O(22′)-Cu(2)-N(8′)
88.86(9)
93.5(1)
95.1(1)
Table 4. Selected Bond Distances (Å) and Bond Angles (deg) for
CuCl2‚4a
91.80(9) O(22′)-Cu(2)-N(11′) 84.69(8)
Cu(1)-Cl(1)
Cu(1)-N(1)
Cu(1)-N(4)
2.296(1) Cu(1)-N(8)
2.188(3) Cu(1)-N(11)
2.166(4)
2.101(3)
2.082(3)
93.5(1)
O(22′)-Cu(2)-N(14′) 178.5(2)
O(22)-Cu(1)-N(11) 85.69(9) N(4)-Cu(l)-N(8)
O(22)-Cu(1)-N(14) 178.4(2) N(4)-Cu(1)-N(11)
93.0(1)
177.5(1)
C1(1)-Cu(1)-N(1)
94.74(9) N(1)-Cu(1)-N(8) 176.0(2)
C1(1)-Cu(1)-N(4) 107.2(1)
N(1)-Cu(1)-N(11) 90.7(1)
C1(1)-Cu(1)-N(8)
89.23(9) N(4)-Cu(1)-N(8) 94.2(1)
1,4,8,11-tetraazacyclotetradecane,53 which has trans-carboxylate
oxygens at the axial sites and four equatorial cyclam nitrogens.
It also differs substantially from the structure of the Cu(II)
complex of 1,4,8,11-tetrakis(carboxymethyl)-1,4,8,11-tetraaza-
cyclotetradecane (TETA), which features two axial nitrogen
donors and a N2O2 equatorial donor set.54 Recently, a copper-
(II) complex of a dibenzo-cyclam derivative was reported which
has a very similar cis-folded ligand conformation as well as
axial bond elongations.55 However, the Jahn-Teller distortions
in this case amount to only 2-3% of the Cu-N and Cu-O
bond lengths.
C1(1)-Cu(1)-N(11) 166.0(1)
N(4)-Cu(1)-N(11) 86.1(1)
N(8)-Cu(1)-N(11) 85.4(1)
N(1)-Cu(1)-N(4)
84.7(1)
In summary, in stark contrast to the three trans as well as
the two cis coordination configurations well-established for
cyclam and its unbridged derivatives,56 all of the copper(II)
complexes reported here feature the respective cross-bridged
ligand in a cis-V-folded complexing geometry. This is a direct
consequence of the ligand design. Five-coordinate complexes
Cu(ClO4)2 ‚5, CuCl2‚5, and CuCl2‚4a‚H2O are all intermediate
between idealized square pyramidal and trigonal bipyramidal
coordination, while pendant-armed complex [Cu‚8‚Na(H2O)-
ClO4]2‚H2O is best described as distorted octahedral.
Figure 9. Coordination geometry at a copper center of the [Cu‚8‚Na-
(H2O)ClO4]2‚H2O structure.
H(26), of a pendant benzyl phenyl group. This Cu(1)‚‚‚H(26)
distance is 2.74 Å. Treating this as a five-coordinate complex,
both the τ-parameter of 0.17 and dihedral δ angle of 17° reveal
a closer relationship to a square pyramidal geometry,51,52 with
N(4) as the apical site. Important bond angles and distances
are tabulated in Table 4.
Conclusion
Short, efficient syntheses of 1,8-ethylene-cross-bridged cy-
clams and pendant-arm derivatives have been described in detail
and the stereochemical basis for the success of the approach
has been discussed. The synthetic strategy relies upon the highly
regioselective kinetic alkylation of a conformationally well-
defined tetracyclic bisaminal derivative of cyclam and a
subsequent double reductive ring expansion to the tetraazabi-
cyclo[6.6.2] system. The parent molecule 5 is readily synthesized
from cyclam via N,N′-dibenzyl cross-bridged cyclam 3a in four
steps in an overall yield of 53%. The ability of the cross-bridged
cyclams to complex copper(II) in their molecular clefts using
four convergent nitrogen lone pairs has been confirmed by the
syntheses and characterization of these copper(II) complexes.
In each case, cross-bridging enforces a folded cis-V ligand
geometry around the metal center, leading to a distorted square
pyramidal coordination sphere. Further, the attachment of
In contrast to the other complexes, the crystal structure of
the pendant-armed ligand complex [Cu‚8‚Na(H2O)ClO4]2‚H2O
(Figure 9) shows that copper is fully enveloped by the bicyclic
ligand and its two pendant arms in a distorted octahedral N4O2
donor set. The structure features two independent but similar
copper macrobicyclic units which form a head-to-tail chain along
the crystallographic x-axis, connected by two types of dimeric
sodium units (see Supporting Information for a structural
depiction showing these interactions). Around each independent
copper core, only one of which is shown in Figure 9, the Jahn-
Teller distortion is manifested in the copper bonding to one axial
carboxylate oxygen, O(1) and O(1′), and one axial tertiary
nitrogen, N(8) and N(8′), respectively. These axial bonds are
significantly elongated, with an average Cu-Oax distance of
2.31 Å and Cu-Nax distance of 2.23 Å, as compared to the
more strongly ligated equatorial copper-donor atom distances
(average Cu-Oeq distance is 2.00 Å; Cu-Neq distance is 2.05
Å). Each copper atom and its respective equatorial donor atoms
are coplanar, with equatorial cis angles at both copper centers
summing to exactly 360.0°. Important bond distances and angles
are presented in Table 5. This coordination geometry differs
from that of the Cu(II) complex of 1,8-bis(carboxymethyl)-
(53) (a) Chapman, J.; Ferguson, G.; Gallagher, J. F.; Jennings, M. C.;
Parker, D. J. Chem. Soc., Dalton Trans. 1992, 345. (b) Chen, L.; Thompson,
L. K.; Bridson, J. N.; Xu, J.; Ni, S.; Guo, R. Can. J. Chem. 1993, 71, 1805.
(c) Belsky, V. K.; Streltsova, N. R.; Kuzmina, E. N.; Nazarenko, A. Y.
Polyhedron 1993, 12, 831.
(54) Moi, M. K.; Yanuck, M.; Deshpande, S. V.; Hope, H.; DeNardo,
S.; Meares, C. F. Inorg. Chem. 1987, 26, 3458.
(55) Rainer, M.; Voet, U. Z. Naturforsch, 1999, 54b, 321.
(56) Bakaj, M.; Zimmer, M. J. Mol. Struct. 1999, 508(1-3), 59.

Cross-Bridged Cyclams and Pendant-Armed DeriVatiVes
J. Am. Chem. Soc., Vol. 122, No. 43, 2000 10569
mol) was added in one portion to a stirred solution of 2 (9.04 g, 40.6
mmol) in MeCN (200 mL), and the reaction mixture was stirred for
14 days. The precipitate was collected by suction filtration, washed
with MeCN (2 × 20 mL) followed by CH2Cl2 (2 × 50 mL), and residual
solvent was removed to give the product as a white powder (21.35 g,
93%): mp 147-151 °C (dec); 1H NMR (D2O) δ 1.86-1.96 (dm, 2H,
J ) 15.2 Hz, C-CHeq-C), 2.20-2.39 (qm, 2H, J ) 15.2 Hz,
C-CHax-C), 2.81 (td, 2H, J ) 12.3, 2.8 Hz), 3.17-3.30 (dm, 4H, J
) 12.3), 3.39-3.65 (m, 6H), 3.74 (td, 2H, J ) 13.2, 3.4 Hz), 4.43 (td,
2H, J ) 13.1, 4.0 Hz), 4.75 and 5.27 (AB, 4H, J ) 13.0 Hz, CH2Ph),
5.08 (s, 2H, CH), 7.40-7.80 (m, 10H); 13C NMR (D2O) δ 18.8, 46.7,
47.5, 51.9, 61.1, 63.1, 77.5, 125.5, 130.2, 132.2, 134.0; IR (KBr) 3051,
3030, 3003, 2995, 2960, 2943, 2869, 2853, 2844, 2812, 1456, 1356,
1137, 1056, 871, 722, 706 cm-1. Anal. Calcd for C26H36N4Br2‚H2O:
C, 54.36; H, 6.67; N, 9.75. Found: C, 54.71; H, 6.79; N, 9.40.
(10br,10cr)-Decahydro-3a,8a-dimethyl-1H,6H-3a,5a,8a,10a-tet-
razapyrenium Diiodide Monohydrate (3b‚H2O).6 Excess MeI (10
mL, 161 mL) was added in one portion to a stirred solution of 2 (2.34
g, 10.5 mmol) in MeCN (75 mL). The reaction mixture was stirred at
room temperature for 72 h in a tightly capped flask, during which time
the product crystallized from the solution. Excess MeI was evaporated
with a stream of N2, and the reaction mixture was concentrated to
approximately one-half volume (35 mL). The white crystalline product
was collected by suction filtration and washed with MeCN (35 mL),
and residual solvent was removed (100 °C, 0.5 mmHg) to yield 5.18
g of crude product. This was recrystallized from MeOH to give 3.99 g
pendant arms to the parent ligand has been shown to result in
full envelopment of the copper(II) cation in a six-coordinate
octahedral embrace. We believe that these ligands and the related
[6.5.2] and [5.5.2] systems will prove highly valuable in
diagnostic and therapeutic biomedical applications and catalytic
chemistry, as well as stereoselective synthesis. We and others
are actively working toward these ends.18,40,57
Experimental Section
Caution! Although we did not encounter any difficulties, metal
perchlorate salts with organic ligands and solVents are potentially
explosiVe and should be prepared and handled in only small quantities
with great care.
1
General Methods. Unless indicated otherwise, H NMR and 13C
NMR spectra were recorded at 360.13 and 90.56 MHz, respectively,
and referenced against internal TMS. MeCN was used as the secondary
internal standard for NMR spectra of D2O solutions; the methyl
resonance was set at δ 2.06 and δ 1.7 for 1H and 13C spectra,
respectively. Resonance assignments are indicated only where they are
firmly established, for example, by coupling and symmetry consider-
ations or additional NMR experiments such as DEPT, COSY, HET-
COR, or DNMR. IR spectra were recorded as KBr pellets on a Nicolet
MX-1 FT spectrophotometer. Electronic spectra were obtained on a
Cary 219 spectrophotometer. X-band EPR spectra were recorded on a
laboratory-assembled spectrometer with a Bruker ER 041 XK-H
microwave bridge. Low-resolution MS and elemental analyses were
obtained from the University of New Hampshire University Instru-
mentation Center. HRMS were obtained from the Dupont Merck
Pharmaceutical Company (Wilmington, DE) or the Midwest Center
for Mass Spectrometry (Lincoln, NE). Melting points are uncorrected.
Reactions were run under nitrogen atmosphere with magnetic stirring.
Bulk solvent removal was by rotary evaporation under reduced pressure
and trace solvent removal from solids was by vacuum pump. Unless
specified otherwise, anhyd Na2SO4 was used to dehydrate organic
solutions.
1
(72%) of pure product: mp ∼220 °C (dec); H NMR (D2O) δ 1.86-
1.96 (dm, 2H, J ) 15.3 Hz, H2eq), 2.31-2.47 (m, 2H, H2ax), 2.74 (td,
2H, J ) 12.4, 3.3 Hz, H1ax), 3.13-3.23 (m, 6H, H1eq and H4eq and
H
5ax), 3.23-3.33 (dm, 2H, J ) 12.5 Hz, H5eq ), 3.36 (s, 6H, CH3), 3.68
(td, 2H, J ) 13.2, 3.8 Hz, H3ax), 3.74-3.83 (dm, 2H, J ) 13.0 Hz,
3eq), 4.47 (td, 2H, J ) 13.7, 5.5 Hz, H4ax), 4.69 (s, 2H, CH); 13C
H
NMR (D2O) δ 19.0 (C2), 46.8 (C5), 48.9 (CH3), 50.3 (C4), 65.7 (C3),
77.2 (CH) (Note: DEPT, 1H COSY and 1H-13C HETCOR experiments
were used to make resonance assignments. See Figure 2 for carbon
numbering.); IR (KBr) 3010, 2990, 2945, 2905, 2840, 2790, 2740, 2665,
1460, 1450, 1380, 1375, 1270, 1105 cm-1. Anal. Calcd for C14H28N4I2‚
H2O: C, 32.08; H, 5.77; N, 10.69. Found: C, 32.43; H, 5.66; N, 10.80.
Note: A run employing 2.05 g of 2, 12 mL of MeI, and 10 mL of
MeCN stirred for 94 h at room temperature yielded 80% of 3b‚H2O
containing only trace impurities, making MeOH recrystallization
unnecessary.
Materials. Solvents were either used as purchased or dried according
to standard literature procedures.58 Benzyl bromide, methyl iodide, ethyl
bromoacetate, R-chloroacetamide, and glyoxal (40 wt % aq soln) were
obtained from the Aldrich Chemical Co. 1,4,8,11-Tetraazacyclotet-
radecane (cyclam, 1) was obtained from the Strem Chemical Co. Copper
salts were obtained from Alfa Aesar.
4,11-Dibenzyl-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (4a). De-
tails of the previously communicated prepn:7 NaBH4 (66.70 g, 1.76
mol) was added in small portions over 1 h to a stirred solution of 3a
monohydrate (20.25 g, 34.77 mmol) in 95% EtOH (900 mL). The
reaction mixture was stirred at room temperature for 16 days. Excess
NaBH4 was then decomposed by slow addition (with cooling) of 3M
HCl (700 mL), and the solvent was removed. The resulting white solid
was dissolved in H2O (400 mL), the pH was adjusted to 14 (KOH
pellets with cooling), and the basic solution was extracted with benzene
(6 × 250 mL). The combined extracts were dried and the solvent was
removed to yield 12.44 g (88%) of 4a (>99% purity by NMR): mp
89-91 °C; 1H NMR (CDCl3) δ 1.31-1.45 (m, 2H, C-CH-C), 1.50-
1.65 (m, 2H, C-CH-C), 2.26-2.53 (m, 12H), 2.38-2.54 (XX′ of
AA′XX′, 2H, NCH2CH2N cross-bridge), 2.85 (ddd, 2H, J ) 13.0, 11.2,
4.0 Hz), 3.11-3.29 (AA′ of AA′XX′, 2H, NCH2CH2N cross-bridge),
3.17 and 3.78 (AX, 4H, J ) 13.5 Hz, CH2Ph), 3.96 (ddd, 2H, J )
12.1, 10.8, 4.3 Hz), 7.15-7.45 (m, 10H); 13C NMR (CDCl3) δ 28.06
(CH2CH2CH2), 52.0, 54.3, 56.6, 57.2, 57.6, 60.0, 126.5, 128.0, 128.9,
141.0; MS, m/z 406 (M+); IR (KBr) 3050, 3025, 3000, 2960, 2945,
2915, 2900, 2860, 2800, 2770, 2720, 2860, 2655, 1445, 1435, 1345,
1093, 710, 660 cm-1. Anal. Calcd for C26H38N4: C, 76.80; H, 9.42; N,
13.78. Found: C, 76.72; H, 9.82; N, 13.82.
cis-Decahydro-1H,6H-3a,5a,8a,10a-tetraazapyrene (2). Modifica-
tion and scale-up of the previously reported prepn:21,59 A solution of 1
(8.47 g, 42.3 mmol) and 7.0 mL of 40 wt % aq glyoxal in 700 mL of
MeCN were stirred at room temperature for 1 h, then at 50-60 °C for
2 h. After the reaction mixture was cooled to room temperature, the
solvent was removed, the resulting orange solid was taken up in CHCl3
(500 mL), the solution was dried and filtered, and CHCl3 was removed
from the filtrate to give a white solid product (Et2O trituration aids
solidification). Sublimation (80 °C, 0.02 mmHg) gave pure 2 (7.65 g,
1
81%): mp 82-84 °C (lit21 82.5-85 °C); H NMR (CDCl3) δ 1.22
(dpapp, 2H, J ) 13.1, 2.6 Hz, H2eq), 2.05-2.17 (tmapp, 2H, J ) 11.4
Hz, H1ax), 2.15-2.17 (m, 4H, H2ax and H5ax), 2.28-2.36 (dm, 2H, J )
10.9 Hz, H4eq), 2.69-2.77 (dm, 2H, J ) 11.0 Hz, H5eq), 2.87-3.03
(m, 6H, H3eq and H3ax and H1eq), 3.08 (s, 2H, N-CH-N), 3.44-3.60
(tmapp, 2H, J ) 10.3 Hz, H4ax); 13C NMR (CDCl3) δ 19.7 (C2), 44.8
(br, C4), 52.6 (br, C3), 54.4 (br, C5), 56.1 (br, C1), 77.1 (NCN) [Note:
1H COSY, 1H-13C HETCOR, NOESY experiments, as well as
connectivities derived from 13C DNMR, were used to make resonance
assignments (see Figure 2 for compound numbering)]; IR (CCl4) 2835,
2855, 2705, 2650 cm-1. Anal. Calcd for C12H22N4: C, 64.83; H, 9.97;
N, 25.20. Found: C, 65.02; H, 10.31; N, 25.12.
(10br,10cr)-Decahydro-3a,8a-bis(phenylmethyl)-1H,6H-3a,5a,-
8a,10a-tetrazapyrenium Dibromide Monohydrate (3a‚H2O). Details
of the previously communicated prepn:7 Benzyl bromide (100 g, 0.585
4,11-Dimethyl-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (4b). Scale-
up and optimization of the previously reported prepn:6 NaBH4 (4.00 g,
105 mmol) was slowly added in small portions to a stirred solution of
3b monohydrate (5.71 g, 10.9 mmol) in 95% EtOH (200 mL). The
reaction mixture was stirred at room temperature for 72 h. Excess
NaBH4 was then decomposed by slow addition of 10% aq HCl (50
mL) (final pH ∼ 1). Absolute EtOH (400 mL) was added and the
(57) Weisman, G. R.; Hill, D. C.; Bist, S.; Condon, J. S.; Fagan, M. A.;
Wong, E. H. Manuscript in preparation.
(58) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon Press: Oxford, 1988.
(59) For an alternative prepn, see ref 39.

10570 J. Am. Chem. Soc., Vol. 122, No. 43, 2000
Wong et al.
mixture was reduced to dryness by rotary evaporation. The white residue
was dissolved in H2O (60 mL), the solution was made strongly basic
(pH ) 14) by slow addition of NaOH pellets with cooling, and the
basic solution was extracted with benzene (4 × 100 mL). The combined
extracts were dried and the solvent was removed to give the crude
product, which was kugelrohr-distilled from KOH pellets (0.02 mmHg,
air bath temp 70-75 °C) to yield 2.49 g (9.80 mmol; 90%) of pure 4b
as a viscous, colorless liquid: 1H NMR (C6D6) δ 1.29-1.50 (m, 4H,
C-CH2-C), 2.12 (s, 6H, CH3), 2.20 (dt, 2H, J ) 11.9, 3.7 Hz), 2.20-
2.31 (m, 4H), 2.36 (ddd, 2H, J ) 13.0, 4.7, 3.0 Hz), 2.45-2.55 (XX′
of AA′XX′, 2H, NCH2CH2N cross-bridge), 2.46-2.54 (m, 2H), 2.71-
2.80 (m, 4H), 3.15-3.25 (AA′ of AA′XX′, 2H, NCH2CH2N cross-
bridge), 3.74 (td, 2H, J ) 11.6, 4.3 Hz); 13C NMR (C6D6) δ 28.5
(CH2CH2CH2), 42.6 (CH3), 51.6, 56.3, 56.4, 57.4, 61.6 (1H COSY and
1H-13C HETCOR 2D experiments are also consistent with the
structure); MS, m/z 254 (M+); IR (KBr) 2945, 2920, 2900, 2820, 2800,
2775, 1445, 1355 cm-1. Anal. Calcd for C14H30N4: C, 66.09; H, 11.89;
N, 22.02. Found: C, 66.15; H, 12.09; N, 22.13.
HRMS exact mass calcd for C20H39N4O4, 399.2971; found, 399.2963.
When the reaction was scaled up to 2.14 g of 5, 3.39 g (90%) of 6
were obtained. Stirring the crude product for 10 min in ice-cold 20%
aq NaOH prior to extraction was necessary in order to avoid recovery
of partially protonated product.
4,11-Bisacetamido-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane Hy-
drate (7‚0.5H2O). A solution of 5 (0.2425 g, 1.071 mmol) in MeCN
(25 mL) was treated with anhyd K2CO3 (0.60 g, 4.3 mmol), KI (0.71,
4.3 mmol) and R-chloroacetamide (0.4233 g, 4.53 mmol), and the
resulting mixture was heated at 60 °C under N2 for 23 h. The solvent
was then removed, the residue was dissolved in H2O (15 mL), and the
solution was made strongly basic (pH 14, slow addition of KOH pellets
with cooling). The resultant solution was extracted with CHCl3 (6 ×
25 mL), the combined organic phases were dried, the solvent was
removed, and the residue was dissolved in EtOH (25 mL). EtOH was
subsequently removed (EtOH addition and removal is necessary in order
to remove CHCl3, which forms a solvate with the product) to afford
1
0.3501 g (94%) of waxy 7: mp 164-165 °C dec; H NMR (CD3CN,
ref CD2HCN set at 1.94) δ 1.39-1.64 (m, 4H, CH2CH2CH2), 2.33-
2.58 (m, 12H), 2.66-2.75 (m, 4H), 2.79 and 3.07 (AB, 4H, J ) 16.1
Hz, NCH2CONH2), 2.97-3.08 (AA′ of AA′XX′, 2H, NCH2CH2N
bridge), 3.99 (ddd, 2H, J ) 12.6, 9.0, 5.2 Hz), 5.76 (br s, 2H, amide
NH), 6.78 (br s, 2H, amide NH); 13C NMR (CD3CN, ref CD3CN set at
1.39) δ 28.3, 54.0, 54.2, 56.9, 58.0, 59.0, 61.0, 175.1; IR (KBr) 3444,
3325, 3248, 3184, 1685, 1651 cm-1; MS (EI) m/z 340.3 (M+); Anal.
Calcd for C16 H32N6O2‚0.5 H2O: C, 54.99; H, 9.52; N, 24.05. Found:
C, 54.90; H, 9.18; N, 23.64.
4,11-Bis-(carboxymethyl)-1,4,8,11-tetraazabicyclo[6.6.2]hexa-de-
cane Tetrahydrochloride Hydrate (H28‚4HCl‚1.5H2O). Compound
6 (0.635 g, 1.593 mmol) was dissolved in 6 N HCl (30 mL), and heated
to 100 °C with stirring under N2 for 48 h. The solvent was removed
under reduced pressure to give the product as an oil which crystallized
after removal of residual solvent (0.777 g, 95%): mp 155-160 °C; 1H
NMR (D2O) δ 1.63-1.89 (dm, 2H, J ) 16.6 Hz), 2.23-2.53 (m, 2H),
2.93-3.05 (tm, 4H, J ) 13.4 Hz), 3.04-3.10 (m, 4H), 3.24-3.43 (m,
4H), 3.45-3.73 (m, 8H), 3.50 and 3.90 (AB, 4H, J ) 17.8 Hz, NCH2-
COOH); 13C NMR (D2O) δ 19.7 (CH2CH2CH2), 47.4, 47.6, 53.3, 55.2,
57.7, 60.0, 172.9; IR (KBr) 3411, 3291, 3177, 2936, 2860, 2714, 2600,
2528, 1709 (CO), 1461, 1429, 1232, 989, 664 cm-1. Anal. Calcd for
C16H30N4O4‚4HCl‚1.5H2O: C, 37.29; H, 7.24; N, 10.87. Found: C,
37.20; H, 7.41; N, 11.21.
1,4,8,11-Tetraazabicyclo[6.6.2]hexadecane (5). Details of the previ-
ously communicated prepn:7 Hydrogenolysis of 4a was carried out in
a glass apparatus designed for exclusion of O2 and for measurement of
H2 uptake with maintenance of constant pressure. 10% Pd/C (0.80 g)
and glacial HOAc (125 mL) were added to a 500 mL hydrogenation
flask which was connected to the apparatus. The system was evacuated
(by means of a water aspirator) and flushed with nitrogen four times.
The system was then evacuated, filled with H2 (759 mmHg), and
catalyst was equilibrated for 1 h until H2 uptake ceased. A solution of
4a (4.54 g, 11.17 mmol) in glacial HOAc (5 mL) was then added, and
the mixture was stirred under H2 (759 mmHg). After 21 h the reaction
was stopped (H2 uptake was 525 mL, 97% of the theoretical value).
The apparatus was evacuated and flushed with nitrogen four times,
the reaction flask was removed from the apparatus, the contents were
filtered through diatomaceous earth, and the catalyst and diatomaceous
earth were washed with glacial acetic acid (3 × 10 mL). The solvent
was removed from the filtrate and washings to give a light yellow oil
which was dissolved in H2O (20 mL). The solution was made strongly
basic (pH 14, KOH pellets with cooling) and was extracted with
benzene (6 × 25 mL). The combined extracts were dried and the solvent
was removed under reduced pressure to give an oil (2.35 g, 93% crude
yield). This oil was dissolved in Et2O (17 mL) and then crystallized at
-78 °C (acetone/dry ice bath) in a fritted Schlenk tube to yield 2.02 g
(80%) of pure white 5 (hygroscopic): mp 48-49 °C; 1H NMR (C6D6)
δ 1.16 (dtt, 2H, J ) 15.1, 6.8, 2.7 Hz, H6,13eq), 1.53 (dtt, 2H, J ) 15.0,
9.0, 2.7 Hz, H6,13ax), 2.07-2.26 (overlapping AA′BB′, 4H and m, 2H,
NCH2CH2N bridge and H2,9 or H3,10), 2.31 (ddd, 2H, J ) 12.9, 6.7, 2.6
Hz, H5,12 or H7,13), 2.42-2.64 (m, 6H, H5,12 and H7,13 and H2,9 or H3,10),
2.74-2.86 (m, 4H, H2,9 and/or H3,10), 3.21 (ddd, 2H, J ) 12.8, 8.8, 2.7
Hz, H5,12 or H7,13), 3.59 (br s, 2H, NH) (Note: 1H COSY was used to
make resonance assignments.); 13C NMR (C6D6) δ 24.9 (CH2CH2CH2),
47.4, 50.7, 52.4, 56.1, 59.0; IR (CHCl3) 3260, 2955, 2925, 2890, 2870,
2825, 1490, 1352, 1248, 1135, 1090, 652 cm-1; HRMS exact mass
calcd for C12H26N4, 226.2157; found, 226.2151.
Attempted Preparation of 4,11-bis-(acetamido)-1,4,8,11-tetra-
azabicyclo[6.6.2]hexadecane (7). Actual Alternative Preparation of
4,11-bis-(carboxymethyl)-1,4,8,11-tetraazabicyclo[6.6.2]hexa-
decane (H28). Compound 6 (69.8 mg, 0.175 mmol) was dissolved in
concd aq NH3 (2 mL). The mixture was heated to 100 °C for 3 days
in a pressure tube. The mixture was cooled for 15 min (0-5 °C), and
solvent was removed under reduced pressure to give an oil which
crystallized after removal of residual solvent (37.2 mg, 62%): mp 270-
1
275 °C (dec); H NMR (D2O) δ 1.70-1.77 (dm, 2H, J ) 16.7 Hz),
2.20-2.45 (dtt, 2H, J ) 16.0, 12.9, 3.2 Hz ), 2.56 (dd, 2H, J ) 13.6,
2.2 Hz), 2.80-3.79 (m, 17H), 3.31 (XX′ of AA′XX′, 2H, NCHCHN),
4.14 (AA′ of AA′XX′, 2H, NCHCHN), 3.69 (td, 2H, J ) 14.5, 3.5
Hz); 13C NMR (D2O) δ 20.51 (CH2CH2CH2), 49.1, 49.6, 52.6, 57.8,
58.8, 59.1, 171.5; MS (CI, NH3), m/z 343 (M + 1); IR (KBr) 3409,
2999, 2991, 2957, 2898, 2838, 2810, 1629, 1485, 1404, 1392, 1101,
1062, 791, 740 cm-1. HRMS exact mass calcd for C16H31N4O4,
343.2345; found, 343.2349.
4,11-Bis-(carboethoxymethyl)-1,4,8,11-tetraazabicyclo[6.6.2]-
hexadecane (6). Anhydrous Na2CO3 (0.78 g, 7.3 mmol) and ethyl
bromoacetate (1.23 g, 7.39 mmol) were added sequentially to a solution
of 5 (0.831 g, 3.67 mmol) in MeCN (30 mL). The mixture was heated
to 50 °C and stirred under N2 for 22.5 h. The solvent was then removed
and the residue was dissolved in 20% aq NaOH (30 mL) at ice bath
temperature (0-5 °C). This solution was extracted with cold (0-5 °C)
CHCl3 (6 × 50 mL), the combined extracts were dried, and the solvent
was removed to give 1.52 g of 6 (essentially quantitative): mp 57-59
°C; 1H NMR (C6D6) δ 0.97 (t, X3 of ABX3, 6H, J ) 7.1 Hz, CH2CH3),
1.19-1.41 (m, 4H, CH2CH2CH2), 2.17 (ddd, 2H, J ) 13.4, 3.7, 1.8
Hz), 2.29 (ddd, 2H, J ) 13.0, 4.7, 2.5 Hz), 2.43-2.57 (XX′ of AA′XX′,
2H, NCHCHN), 2.55-2.90 (m, 10H), 3.08 and 3.20 (AB, 4H, J )
16.7 Hz, NCH2COOEt), 3.21-3.35 (AA′ of AA′XX′, 2H, NCHCHN),
3.69 (td, 2H, J ) 11.9, 4.4 Hz), 3.96 and 3.97 (AB of ABX3, 4H, J )
7.1 Hz, CH2CH3); 13C NMR (C6D6) δ 14.4 (CH2CH3), 28.2 (CH2CH2-
CH2), 50.9, 53.3, 54.7, 56.4, 57.3, 59.8, 60.2, 171.6; IR (KBr) 2969,
2945, 2926, 2907, 2851, 2833, 2819, 2784, 1743 (CO), 1461, 1444,
1379, 1365, 1301, 1184, 1162, 1137, 1124 cm-1; MS, m/z 398 (M+);
(10br,10cr)-Decahydro-3a-(phenylmethyl)-1H,6H-3a,5a,8a,10a-
tetraazapyrenium Bromide Hydrate (9a‚1.5H2O).29,30 Benzyl bromide
(0.79 g, 4.62 mmol) was added in one portion to a stirred solution of
2 (0.75 g, 3.37 mmol) in toluene (25 mL) at room temperature. The
reaction mixture was stirred for 11 days. Precipitated product was
collected by suction filtration and was washed with toluene (3 × 15
mL). Residual solvent was removed under vacuum to give 0.82 g of
white powder (62%) (additional pure product slowly crystallized from
the filtrate): mp 137-140 °C (dec); 1H NMR (D2O) δ 1.40-1.51 (dm,
1H, J ) 14.1 Hz, H6 or 13-eq), 1.70-1.83 (dm, 1H, J ) 14.9 Hz, H6 or
13-eq), 2.09-2.34 (m, 3H), 2.41-2.53 (m, 2H), 2.65 (td, 1H, J ) 12.2,
2.8 Hz), 2.92-3.18 (m, 7H), 3.20-3.27 (dm, 1H, J ) 13.6 Hz), 3.28-
3.36 (dm, 1H, J ) 13.7 Hz), 3.43-3.61 (m, 2H), 3.68 (br s, 1H, CH,

Cross-Bridged Cyclams and Pendant-Armed DeriVatiVes
Table 6. X-ray Crystal Data Collection and Refinement Details
J. Am. Chem. Soc., Vol. 122, No. 43, 2000 10571
3a
9a
Cu(ClO4)2‚5
CuCl2‚5‚3H2O
CuCl2‚4a‚H2O
[Cu‚8‚Na(H2O)ClO4]2‚H2O
formula
fw
space group
a (Å)
b (Å)
c (Å)
C26H36Br2N4
564.41
P21/c
10.261(5)
16.513(4)
15.821(3)
90
98.19(3)
90
2653.3(14)
4
C19H33BrN4O2
429.40
P21
9.1396(2)
13.1846(3)
9.1505(2)
90
114.7825(8)
90
1001.11(6)
2
C12H26Cl2CuN4O8
488.81
P212121
9.295(2)
11.418(2)
17.598(3)
90
90
90
1867.9(8)
4
1.738
C12H32Cl2CuN4O3
414.86
Pna21
25.458(2)
7.659(1)
9.435(1)
90
90
90
1839.7
4
1.498
C26H40Cl2CuN4O
559.08
P1
10.079(1)
14.487(1)
9.869(1)
97.751(6)
111.072(4)
98.433(5)
1302.4
C32H62Cl2Cu2N8Na2O19
1106.85
P21/c
16.155(1)
9.377(1)
29.023(1)
90
91.494(2)
90
4395.1
4
1.673
221
0.71073
0.033 (R)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
2
F
calc (g‚cm-3
)
1.413
1.425
173(2)
0.71073
0.0474
1.426
294
0.71073
0.038 (R)
T (K)
250(2)
0.71073
0.0623
256(2)
0.71073
0.0465
219
0.71073
0.029 (R)
wavelength (Å)
R1 indices
[I > 2σ (I)]
wR2 indices
(all data)
0.1798
0.1603
0.1265
0.027 (Rw)
0.035(Rw)
0.033 (Rw)
broadened by gauche coupling), 4.19 (td, 1H, J ) 13.1, 3.5 Hz), 4.36
(br s, 1H, CH, broadened by gauche coupling), 4.77 and 5.05 (AB,
2H, J ) 13.3 Hz, CH2Ar), 7.45-7.68 (m, 5H); 13C NMR (D2O) δ
18.8 (CH2CH2CH2), 19.1 (CH2CH2CH2), 42.7, 47.4, 49.3, 52.1, 52.7,
54.0, 54.7, 60.6, 63.3, 70.3, 82.6, 126.4, 130.1, 131.8, 134.1; IR (KBr)
3063, 3034, 3025, 3006, 2947, 2931, 2901, 2861, 2838, 2793, 2772,
1651, 1457, 1357, 1142, 1110, 891, 707 cm-1. Anal. Calcd for C19H29N4-
Br‚1.5H2O: C, 54.28; H, 7.67; N, 13.32. Found: C, 53.95; H, 7.81;
N, 13.13.
mmol) was added. The resulting clear blue solution was refluxed for 2
h, cooled, and evaporated to dryness. The blue residue was taken up
in a minimal amount of methanol and filtered through diatomaceous
earth. The filtrate was evaporated to dryness, and the residue was
recrystallized from 95% ethanol/ether. The deposited blue crystals were
collected and dried (111 mg, 61% yield). Anal. Calcd for C16H28ClCuN4-
NaO8‚1.5H2O: C, 34.72; H, 5.65; N, 10.12. Found: C, 34.45; H, 5.47;
N, 9.90. Electronic data: (KBr) λmax, 629 nm; (MeOH) λmax, 642 nm
(ꢀ ) 20 M-1 cm-1).
Cu(ClO4)2‚5. An amount of Cu(ClO4)2‚6H2O (160 mg, 0.43 mmol)
and ligand 5 (100 mg, 0.44 mmol) in 10 mL of absolute methanol was
refluxed for 2 h. The indigo solution was cooled to room temperature
and filtered through diatomaceous earth. The filtrate was evaporated
to dryness and recrystallized from 95% aq. ethanol/ether to give 135
mg (63% yield) of dark-blue crystals. Anal. Calcd for C12H26Cl2CuN4O8‚
0.5H2O: C, 28.95; H, 5.47; N, 11.25. Found: C, 29.13; H, 5.99; N,
11.19. UV-vis: (KBr) λmax, 582 nm; (MeCN): λmax, 567 nm (ꢀ )
120 M-1 cm-1). Dark-blue X-ray-quality single crystals were obtained
by slow evaporation from a 95% ethanol/ether solution.
Cu(ClO4)2‚7. Compound 7 (39.9 mg, 0.117 mmol) was dissolved
in EtOH (5 mL), Cu(ClO4)2 (46.2 mg, 0.125 mmol) was added to this
solution, and the resulting mixture was heated at reflux for 4 h under
N2. Upon cooling to room temperature, two different materials
precipitated. One was light blue in color and fluffy; the other was dark
blue in color and granular. These two materials could be attributed to
a complex (dark precipitate) and a polymer (light blue precipitate).
EtOH (5 mL) was added and the heating was continued for 2.5 h. The
reaction mixture was cooled to room temperature and the supernatant
was removed by pipet and filtered through a glass wool plug in a pipet.
The filtrate was placed in a closed chamber designed to allow slow
diffusion of Et2O into the solution. After 24 h, a granular solid had
precipitated. This solid was dissolved in 95% EtOH (10 mL).
Approximately 3 mL of this solution was diluted with 95% EtOH (9
mL) and this solution was put in the Et2O diffusion chamber. After 5
days, crystals had formed, but these crystals were not suitable for X-ray
crystallography: IR (KBr) 1665 cm-1 (CdO); UV-vis: (MeOH, 2.2
× 10-3 M) λmax, 630 nm (ꢀ ) 24 M-1 cm-1). Anal. Calcd for C16H32N6-
CuCl2O10: C, 31.87; H, 5.35; N, 13.94. Found: C, 31.65; H, 5.19; N,
13.71. Other crystallization attempts with EtOH, 95% EtOH, and
CH3CN with Et2O diffusion techniques were unsuccessful in affording
X-ray quality crystals.
X-ray Diffraction Studies. Crystal, data collection, and refinement
parameters of the six X-ray diffraction studies are given in Table 6.
Additional details, including full atomic coordinates, complete bond
lengths and angles, etc., are included in the Supporting Information.
All software and sources of the scattering factors are contained in the
SHELXTL (5.03) and (5.10) program library.60 3a: Crystals suitable
for crystallographic structural determination were obtained by allowing
the MeCN filtrate from a benzylation of 2 to evaporate slowly over a
period of 18 days to give colorless blocks of anhydrous 3a. The
systematic absences in the diffraction data are uniquely consistent with
the space group P 21/c which was supported by the chemically
reasonable and computationally stable results of refinement. The
structure was solved by direct methods, completed by subsequent
difference Fourier syntheses and refined by full-matrix least-squares
procedures. All non-hydrogen atoms were refined with anisotropic
thermal parameters, and hydrogen atoms were treated as idealized
contributions. The two bromide counterions were disordered over two
CuCl2‚5. A solution of 200 mg (0.88 mmol) of ligand 5 and 150
mg (0.88 mmol) of CuCl2‚2H2O was refluxed in 10 mL of 95% aq
ethanol for 2 h. The resulting dark-blue solution was filtered through
diatomaceous earth and the filtrate was subjected to ether diffusion,
whereupon dark-blue blocklike crystals were formed (238 mg, 71%).
Anal. Calcd for C12H26Cl2CuN4‚H2O: C, 38.05; H, 7.45; N, 14.79.
Found: C, 38.39; H, 7.68; N, 14.94. UV-vis: (KBr) λmax, 579 nm;
(MeCN) λmax, 596 nm (ꢀ ) 196 M-1 cm-1). X-ray-quality single crystals
were obtained by slow ether diffusion into a 95% aq. ethanol solution.
Cu(ClO4)2‚4b. A solution of Cu(ClO4)2‚6H2O (0.17 g, 0.46 mmol)
and ligand 4b (95 mg, 0.37 mmol) in methanol was refluxed for 2 h.
The resulting dark-blue solution was cooled and subjected to slow ether
diffusion. A 160 mg, 84% yield of the crystalline blue complex was
obtained. An analytical sample was prepared by slow diffusion of a
1:1 benzene:hexane solution into a solution of the complex in methanol.
This gave dark-blue crystals, which were finely ground and dried for
elemental analysis. Anal. Calcd for C14H30Cl2CuN4O8: C, 31.44; H,
6.03; N, 10.47. Found: C, 31.75; H, 6.22; N, 10.39. UV-vis: (KBr)
λ
max, 625 nm; (MeCN) λmax, 302 nm (ꢀ ) 5730 M-1 cm-1), 618 nm (ꢀ
) 76 M-1 cm-1).
CuCl2‚4a. A mixture of ligand 4a (77 mg, 0.19 mmol) and CuCl2‚
2H2O (30 mg, 0.18 mmol) in 5 mL of methanol was refluxed for 2 h.
The dark-green solution was cooled and filtered through diatomaceous
earth. Ether diffusion into the filtrate yielded green crystals of the
complex (68 mg, 68% yield). Anal. Calcd for C26H38Cl2CuN4‚H2O:
C, 55.86; H, 7.21; N, 10.02. Found: C, 55.80; H, 7.19; N, 9.95.
Electronic data: (KBr) λmax, 675 nm; (MeCN) λmax, 680 nm (ꢀ 75).
X-ray crystals were grown by slow evaporation of a MeCN solution.
Cu‚8. A solution of H28‚4HCl‚1.5H2O (171 mg, 0.33 mmol) and
Cu(ClO4)2‚6H2O (123 mg, 0.33 mmol) in 5 mL of 95% ethanol was
prepared. An aq solution (1.92 mL; 1.04 N) of sodium hydroxide (1.99
(60) Sheldrick, G. Siemens XRD, Madison, WI.

10572 J. Am. Chem. Soc., Vol. 122, No. 43, 2000
Wong et al.
positions, 90/10 and 85/15, respectively. 9a: Crystals suitable for
crystallographic structural determination were obtained by slow re-
crystallization of 0.03 g of 9a from acetone (10 mL). Systematic
absences were consistent with possible space groups P21 or P21/m. The
E-statistics for the data suggested the chiral space group P21. The
structure was solved using direct methods and atoms located by
difference Fourier synthesis. Refinement was by full-matrix least-
squares fitting. Non-hydrogen atoms were refined anisotropically while
hydrogen atoms were added at calculated positions and treated as
isotropic contributions with thermal parameters defined as 1.2 or 1.5
times that of the parent atom. Cu(ClO4)2‚5: The systematic absences
in the diffraction data were uniquely consistent for the orthorhombic
space group P212121, which yielded chemically reasonable and com-
putationally stable results of refinement. The structure was solved by
direct methods, completed by subsequent difference Fourier syntheses,
and refined by full-matrix least-squares procedures. The absolute
structure was confirmed [Flack parameter ) -0.02(4)]. All non-
hydrogen atoms were refined with anisotropic displacement coefficients.
Amine hydrogens were located in the difference map, and the N-H
bond distance was fixed to 0.9 Å. All other hydrogens were treated as
idealized contributions. CuCl2‚5: The structure was solved by direct
methods (SHELXS). Hydrogen atoms were idealized with C-H ) 0.95
Å. Refinement was by full-matrix least-squares on F. All non-hydrogen
atoms were refined anisotropically, while all H’s were fixed. 198
parameters; data/parameter ratio ) 5.44; final R ) 0.029; Rw ) 0.027;
error of fit ) 0.74; max ∆/σ ) 0.04; largest residual density ) 0.42
e/Å3 near the water molecules. CuCl2‚4a‚H2O: The structure was solved
by automated Patterson analysis (PHASE), and hydrogen atoms were
idealized with C-H ) 0.95 Å. The water hydrogens were calculated
in hydrogen-bonding positions with the chlorides. Refinement was by
full-matrix least-squares on F. All non-hydrogen atoms were refined
anisotropically, while all H’s were fixed. 307 parameters; data/parameter
ratio ) 7.69; final R ) 0.038; Rw ) 0.035; error of fit ) 0.86; max
∆/σ ) 0.01; largest residual density ) 0.29e/Å3, between N(8) and
C(7). [Cu‚8‚Na(H2O)ClO4]2‚H2O: Structure was solved by direct
methods (MULTAN), hydrogen atoms were idealized with C-H )
0.95 Å. The water hydrogens were successfully located and refined.
Refinement by full-matrix least-squares on F, scattering factors included
anomalous terms for Cu, Cl, and Na. All non-hydrogen atoms were
refined anisotropically, H’s were either refined isotropically or fixed.
610 parameters; data/parameter ratio ) 8.11; final R ) 0.033; Rw
)
0.033; error of fit ) 1.02; max ∆/σ ) 0.00; largest residual density )
0.63e/Å3, near Cl2, O11′, and O14′.
Acknowledgment. We thank the Petroleum Research Fund,
administered by the American Chemical Society (21493-AC1);
the DuPont Merck Pharmaceutical Co.; and the National
Institutes of Health (Grant 1 R15 GM55916-01) for support of
this research. EPR data were kindly recorded by Prof. N. Dennis
Chasteen and Dr. Xiaoke Yang. We also thank Dr. Thomas
Harris (DuPont Merck) for helpful discussions and Mr. Steven
Hutchins for NMR experiments on 2. G.R.W. thanks Prof. Roger
Alder (Bristol) for encouragement, stimulating discussions, and
the free exchange of information.
1
Supporting Information Available: 360 MHz H and 90
MHz 13C NMR spectra of 5, 6, 7, and H28 (PDF). Tables of
crystal data, structure solution and refinement, atomic coordi-
nates, bond lengths and angles, and anisotropic thermal param-
eters for 3a, 9a, Cu(ClO4)2‚5, CuCl2‚5, CuCl2‚4a, and [Cu‚8‚
Na(H2O)ClO4]2, as well as structural drawings for Cu(ClO4)2‚
5 and [Cu‚8‚Na(H2O)ClO4]2 showing H-bonding (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA001295J