Ligand influence over the formation of dinuclear [2+2] versus trinuclear [3+3] CuI Schiff base macrocyclic complexes

Article abstract of DOI:10.1021/ic102185y

The preparation and characterization of three new macrocyclic ligands with pendant arms based on the [2+2] condensation of isophthalaldehyde and the corresponding triamine substituted at the central N-atom is reported. None of these new macrocyclic ligands undergo any equilibrium reaction, based on imine hydrolysis to generate [1+1] macrocyclic formation or higher oligomeric compounds, such as [3+3], [4+4], etc., at least within the time scale of days. This indicates the stability of the newly generated imine bond. In sharp contrast, the reaction of the [2+2] macrocyclic Schiff bases with Cu^I generates the corresponding dinuclear Cu^I complexes [Cu ₂(L¹)]²⁺, 1²⁺; [Cu ₂(L²)(CH₃CN)₂]²⁺, 2 ²⁺; and [Cu₂(L³)(CH₃CN) ₂]²⁺, 3²⁺, together with their trinuclear Cu^I homologues [Cu₃(L⁴)]³⁺, 4 ³⁺; [Cu₃(L⁵)(CH₃CN) ₃]³⁺, 5³⁺; and [Cu₃(L ⁶)(CH₃CN)₃]³⁺, 6³⁺, where the [2+2] ligand has undergone an expansion to the corresponding [3+3] Schiff base that is denoted as L⁴, L⁵, or L⁶. The conditions under which the dinuclear and trinuclear complexes are formed were analyzed in terms of solvent dependence and synthetic pathways. The new complexes are characterized in solution by NMR, UV-vis, and MS spectroscopy and in the solid state by X-ray diffraction analysis and IR spectroscopy. For the particular case of the L² ligand, MS spectroscopy is also used to monitor the metal assisted transformation where the dinuclear complex 2 ²⁺ is transformed into the trinuclear complex 5³⁺. The Cu^I complexes described here, in general, react slowly (within the time scale of days) with molecular oxygen, except for the ones containing the phenolic ligands 2²⁺ and 5³⁺ that react a bit faster.

Full text of DOI:10.1021/ic102185y

ARTICLE
pubs.acs.org/IC
Ligand Influence over the Formation of Dinuclear [2+2] versus
Trinuclear [3+3] CuI Schiff Base Macrocyclic Complexes
Arnau Arbuse,† Sukanta Mandal,^ Somnath Maji,^ Ma Angeles Martínez,*,† Xavier Fontrodona,‡ Diana Utz,||
Frank W. Heinemann,§ Sandra Kisslinger,|| Siegfried Schindler,*,|| Xavier Sala,# and Antoni Llobet*,^,#
†Departament de Química and ‡Serveis Científico-Tꢀecnics, Universitat de Girona, Campus de Montilivi, E-17071 Girona, Spain
§Department Chemie und Pharmazie, Anorganische Chemie, Friedrich-Alexander-Universit€at Erlangen-N€urnberg, Egerlandstrasse 1,
91058 Erlangen, Germany
Institut f€ur Anorganische und Analytische Chemie, Justus-Liebig-Universit€at, Giessen, Germany
^Institute of Chemical Research of Catalonia (ICIQ), Av. Països Catalans 16, E-43007 Tarragona, Spain
#Departament de Química, Universitat Autꢀonoma de Barcelona, Ceranyola del Vallꢀes, E-0194 Barcelona, Spain
S Supporting Information
b
ABSTRACT:
The preparation and characterization of three new macrocyclic ligands with pendant arms based on the [2+2] condensation of
isophthalaldehyde and the corresponding triamine substituted at the central N-atom is reported. None of these new macrocyclic
ligands undergo any equilibrium reaction, based on imine hydrolysis to generate [1+1] macrocyclic formation or higher oligomeric
compounds, such as [3+3], [4+4], etc., at least within the time scale of days. This indicates the stability of the newly generated imine
bond. In sharp contrast, the reaction of the [2+2] macrocyclic Schiff bases with CuI generates the corresponding dinuclear CuI
complexes [Cu2(L1)]2+, 12+; [Cu2(L2)(CH3CN)2]2+, 22+; and [Cu2(L3)(CH3CN)2]2+, 32+, together with their trinuclear CuI
homologues [Cu3(L4)]3+, 43+; [Cu3(L5)(CH3CN)3]3+, 53+; and [Cu3(L6)(CH3CN)3]3+, 63+, where the [2+2] ligand has
undergone an expansion to the corresponding [3+3] Schiff base that is denoted as L4, L5, or L6. The conditions under which
the dinuclear and trinuclear complexes are formed were analyzed in terms of solvent dependence and synthetic pathways. The new
complexes are characterized in solution by NMR, UVꢀvis, and MS spectroscopy and in the solid state by X-ray diffraction analysis
and IR spectroscopy. For the particular case of the L2 ligand, MS spectroscopy is also used to monitor the metal assisted
transformation where the dinuclear complex 22+ is transformed into the trinuclear complex 53+. The CuI complexes described here,
in general, react slowly (within the time scale of days) with molecular oxygen, except for the ones containing the phenolic ligands 22+
and 53+ that react a bit faster.
’ INTRODUCTION
their relative position, the number and size of the chelating rings
formed, and the flexibility and shape of the coordinating moiety
on the selective binding of charged or neutral species.8ꢀ10 In
addition, Schiff base macrocyclic ligands can be used as starting
materials to generate the corresponding secondary amines that,
Schiff bases and their related transition metal complexes have
been extensively employed in many fields of science, including
biochemistry, material science, catalysis, supramolecular chem-
istry, transport and separation phenomena, medicine, etc., be-
cause of their synthetic versatility.1ꢀ7 A large variety of [1+1] and
[2+2] macrocyclic ligands have been synthesized in order to
understand the role of the different donor atoms, the influence of
Received: October 29, 2010
Published: July 06, 2011
r
2011 American Chemical Society
6878
dx.doi.org/10.1021/ic102185y Inorg. Chem. 2011, 50, 6878–6889
|

Inorganic Chemistry
ARTICLE
Scheme 1. (A) Synthetic Strategy for the Preparation of Substituted N2-Triamines: H2NC2py, H2NC2PhOH, and H2NC2Et. (B)
Macrocyclic Ligands Obtained from the [2+2] Condensation of N2-Triamines and Isophthalaldehyde, Including Proton Labeling
Used
in turn, can be further functionalized, generating the corresponding
tertiary amines.11 Furthermore, higher condensation products such
as [3+3] and [4+4] have also been reported, although they are
unusual.12 The combination of all these macrocycles provides a
wide family of ligands, which allows an understanding at a
molecular level of phenomena such as anion recognition.13 Further-
more, these ligands can be coordinated to transition metal ions, and
thus, they generate a large family of complexes with subtle
differences that enable the understanding of important phenomena
related to complexꢀDNA interactions14 and the activation of small
molecules, such as dioxygen,15 carbon dioxide, etc.16 Macrocyclic
ligands with peripheral functionalities constitute a specific class
within this type of ligand because they have complementary
properties that can be used for multirecognition processes, specific
separation and transport processes across membranes, or the
additional control of small molecule activation and catalysis.17,18
The use of a metal ion template is a powerful synthetic tool to
direct the Schiff base synthesis to a desired oligomer, controlling
6879
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
the size and shape of the resulting macrocycle.12e,g,i,k In general,
the synthetic routes reported thus far generate single oligomers,
although a few exceptions have been described, particularly for
the discrimination of [2+2] vs. [4+4].12a,d,f
H
meta), 7.38ꢀ7.42 (m, 4H, Hβ0 + Hγ), 7.84ꢀ7.87 (m, 4H, Hortho), 8.08
(s, 4H, ꢀCHdN), 8.49 (d, J = 3.82 Hz, 2H, HR). 13C NMR (CDCl3,
400 MHz)
δ
(ppm): 55.22 (pyNꢀCH2ꢀCH2ꢀNd), 59.72
(pyNꢀCH2ꢀCH2ꢀNd), 61.38 (NꢀCH2ꢀpy), 121.78 (Cβ), 123.11
(Cβ0), 128.70 (Cortho), 128.90 (Cγ), 129.75 (Cortho ), 136.12 (Cmeta),
0
Hereon, we report the synthesis of three new macrocyclic
Schiff base ligands with different pendant arms (2-methylpyridyl,
2-methylphenol, and donor-free ethyl) obtained from the [2+2]
condensation of isophthalaldehyde and N2-functionalized tri-
amine. The new ligands are labeled bsm2py (L1), bsm2PhOH
(L2), and bsm2Et (L3), where “bs” refers to Schiff base, “m” refers
to the meta substitution at the aromatic ring, “2” refers to the
number of methylenic units linking the aminic atoms, and, finally,
“py”, “PhOH,” and “Et” refer to 2-methylpyridyl, 2-methylphe-
nol, and ethyl groups, respectively. The latter groups bonded at
the central N-atom of triamine become the pendant arms of the
macrocyclic ligand. Scheme 1 presents a drawing of these ligands
as well as the synthetic strategy used to obtain them. The
coordination chemistry of these ligands with CuI is also reported,
giving the formation of dinuclear [2+2] and trinuclear [3+3] CuI
complexes. Their interconversion is studied by means of MS
spectroscopy. Corresponding [3+3] Schiff bases are denoted as
L4, L5, and L6.
136.67 (Cqarom), 148.80 (CR), 160.15 (CqR), 161.09 (CHdN). FT-IR
ν (cmꢀ1): 2838 (CꢀH), 1644 (CdN), 1588, 1568 (CꢀC py), 1473,
1433 (CꢀH), 797 (CꢀH ar), 756 (CꢀH py), 692 (CꢀH ar), 614
(CꢀH py). HRMS (m/z): [M + Na]+, 607.3279 (100%); calcd mass,
607.3274.
ftN-C2NPhOH. Salicylaldehyde (3.0 mL, 28 mmol) was added to a
mixture of ftN-C2NH (10.000 g, 28 mmol) in 1,2-dichloroethane
(150 mL). The crude product was stirred for several minutes. Afterward,
NaBH(OAc)3 (8.60 g, 0.039 mol) was slowly added, and the mixture
was stirred at room temperature for 24 h. The organic layer was extracted
after adding 100 mL of water. The aqueous phase was washed with
dichloromethane (2 ꢁ 100 mL). The organic fractions were dried over
MgSO4 and concentrated up to ∼10 mL. Methanol (150 mL) was then
added with stirring, producing the white solid precipitated product,
which was filtered and dried under vacuum. Yield: 11.21 g (87%). Anal.
Calcd (%) for C27H23N3O5 0.25H2O (MW = 473.99 g molꢀ1): C,
3
3
68.42; H, 5.00; N, 8.87. Found: C, 68.33; H, 5.02; N, 9.01. 1H NMR (400
MHz, CDCl3) δ (ppm): 2.95 (t, J = 6 Hz, 4H, ftNꢀCH2ꢀCH2ꢀN),
3.89 (m, 6H, ftNꢀCH2ꢀCH2ꢀN + NꢀCH2ꢀPhOH), 6.24ꢀ6.29 (m,
1H, HR), 6.71ꢀ6.78 (m, 1H, Hβ), 6.92ꢀ6.98 (m, 1H, Hβ0), 7.01ꢀ7.08
(m, 1H, Hγ), 7.70ꢀ7.84 (m, 8H, Har), 9.07 (s, br, PhOH). 13C NMR
(100 MHz, CDCl3) δ (ppm): 34.8 (ftNꢀCH2ꢀCH2ꢀN), 51.5
(ftNꢀCH2ꢀCH2ꢀN), 58.1 (NꢀCH2ꢀPhOH), 116.0 (CRPhOH),
119.5 (CγPhOH), 121.4 (CqRPhOH), 123.2 (Carom) 129.0, 129.1
(CβPhOH, Cβ0PhOH), 132.2 (Cqarom), 133.8 (Carom), 156.8 (CꢀOH),
168.2 (CdO). FT-IR ν (cmꢀ1): 1700 (CdO), 1398 (COꢀN), 754
(CꢀH PhOH), 708 (CꢀH ar), 532 (CꢀH ft). HRMS (m/z): [M +
Na]+, 492.1526 (100%); calcd mass, 492.1535.
’ EXPERIMENTAL SECTION
Physical Methods. IR spectra of solid samples were taken in a
Mattson-Galaxy Satellite FT-IR spectrophotometer using a MKII Gold-
en Gate single reflection ATR system. HRMS analyses were recorded on
a Waters LCT Premier liquid chromatograph coupled time-of-flight
mass spectrometer (HPLC/MS-TOF) with electrospray ionization
(ESI). MS analyses were recorded on an esquire 6000 ESI ion trap
LC/MS (Bruker Daltonics) equipped with an electrospray ion source.
NMR spectra were measured using a Bruker DPX 200 MHz, a Bruker
DXP 300 MHz, or a Bruker DRX 400 MHz instrument. Elemental
analysis was performed using a CHNS-O EA-1108 elemental ana-
lyzer from Fisons. UVꢀvis spectra were taken in a Cary 50 Scan
spectrophotometer.
Materials and Synthesis. All reagents used in the present work
were obtained from Aldrich Chemical Co. and were used without further
purification unless otherwise stated. ftN-C2NH, ftN-C2Npy, and H2N-
C2Npy (see Scheme 1 for abbreviations; H2N-C2Npy is also abbre-
viated as apme18) were synthesized as described in the literature.19 H2N-
C2NEt was synthesized by the following two methods: (i) Method A as
described by Song et al.20 and (ii) a newly developed Method B (see
below). Solvents were purchased from SDS, and they were purified and
dried either by passing them through an activated alumina purification
system (MBraun SPS-800) or by conventional distillation techniques.
Preparation and manipulation of CuI complexes were carried out in a
drybox (MBraun, N2, or Ar) with O2 and H2O concentrations <
1.0 ppm.
H2N-C2NPhOH. Hydrazine monohydrate (9.7 mL, 0.2 mol) was
added to a solution of ftN-C2NPhOH (8.65 g, 18.42 mmol) in chloro-
form/ethanol (60:320 mL). The mixture was stirred at room tempera-
ture for 24 h, and the obtained white precipitate was filtered off and
discarded. The resulting transparent solution was evaporated under
reduced pressure. Chloroform (150 mL) was then added to the residue,
and the mixture was stirred for another 24 h and filtered again.
Evaporation of the chloroform fraction afforded the desired product
as oil. Yield: 2.62 g (85%). Anal. Calcd (%) for C11H19N3O 0.14CHCl3
3
(MW = 226.00 g molꢀ1): C, 59.20; H, 8.54; N, 18.59. Found: C, 59.39;
3
H, 8.81; N, 18.21. 1H NMR (400 MHz, CDCl3) δ (ppm): 2.59 (t, J = 6
Hz, 4H, H2NꢀCH2ꢀCH2ꢀN), 2.86 (t, J = 6 Hz, 4H, H2Nꢀ
CH2ꢀCH2ꢀN), 3.36 (s, br, ꢀNH2), 3.73 (s, 2H, NꢀCH2ꢀPhOH),
6.69ꢀ7.22 (m, 4H, HPhOH). 13C NMR (100 MHz, CDCl3) δ (ppm):
39.3 (ꢀCH2ꢀNH2), 55.7 (ꢀCH2ꢀCH2ꢀNH2), 57.9 (ꢀCH2ꢀ
PhOH), 116.4 (CR), 119.1 (Cγ), 122.7 (CqR), 128.8 (Cβ), 129.3
(Cβ0), 157.5 (CꢀOH). FT-IR ν (cmꢀ1): 1587 (CdC), 1472, 1446
(ꢀCH2ꢀ), 1269, 1256 (ArCꢀOH), 753 (CꢀH ar), 708 (CꢀH
PhOH), 532 (CꢀH ar). ESI-MS (m/z): [M + H]+, 210.1 (100%).
bsm2PhOH (L2). A solution of isophthalaldehyde (0.655 g, 4.88
mmol) in MeCN (50 mL) was slowly added (9 mL/h via syringe
pump) to a solution of H2NꢀC2NPhOH (1.021 g, 4.88 mmol) in
MeCN (50 mL) with stirring. After the mixture was stirred for 24 h, a
white solid was obtained, filtered, and then dried under a vacuum. Yield:
0.797 g (53%). Crystals for X-ray diffraction were obtained by dissolving
0.015 g of L2 in 1 mL of chloroform and then diluting the solution with
MeOH. Slow evaporation of the solvents afforded white crystals suitable
for X-ray diffraction (see the Supporting Information). Anal. Calcd (%)
Ligand Synthesis. bsm2py (L1). A solution of isophthalaldehyde
(0.228 g, 1.70 mmol) in acetonitrile (40 mL) was slowly added (6.0 mL/
h via syringe pump) to a solution of H2N-C2Npy (0.330 g, 1.70 mmol)
in acetonitrile (40 mL) at 0 ꢀC and allowed to react overnight at room
temperature. It was then filtered to remove some solid particles, and the
filtrate was then concentrated in the rotary evaporator, leading to the
separation of an oil. The solvent was decanted, and the oil was dried
under a vacuum. Yield: 0.380 g (76%). Anal. Calcd (%) for C36H40N8
3
0.6H2O 0.4CH3CN (MW = 611.99 g molꢀ1): C, 72.22; H, 6.98; N,
3
3
19.23. Found: C, 72.24; H, 6.45; N, 19.21. 1H NMR (400 MHz, CDCl3)
δ (ppm): 2.98 (t, J = 4 Hz, 8H, pyNꢀCH2ꢀCH2ꢀNd), 3.72 (t, J = 4
Hz, 8H, pyNꢀCH2ꢀCH2ꢀNd), 3.87 (s, 4H, NꢀCH2ꢀpy),
for C38H42N6O2 (MW = 614.78 g molꢀ1): C, 74.24; H, 6.89; N, 13.67.
3
1
Found: C, 73.88; H, 6.84; N, 13.67. H NMR (400 MHz, CDCl3) δ
0
7.04ꢀ7.07 (m, 2H, Hβ), 7.16 (s, 2H, Hortho ), 7.29ꢀ7.33 (m, 2H,
(ppm): 2.98 (t, J = 6 Hz, 8H, CHdNꢀCH2ꢀCH2ꢀN), 3.72 (t, J = 6 Hz,
6880
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
8H, CHdNꢀCH2ꢀCH2ꢀN), 3.90 (s, 4H, NꢀCH2ꢀPhOH),
Method B. Hydrazine monohydrate (6.98 mL, 0.14 mol) was added
to a solution of ftN-C2NEt (5.000 g, 12.77 mmol) in chloroform/
ethanol (50:280 mL). The mixture was stirred at room temperature for
24 h, and then, the obtained white precipitate was filtered off and
discarded. The resulting transparent solution was evaporated under
reduced pressure. Chloroform (150 mL) was then added to the residue,
and the mixture was stirred for another 24 h and filtered again.
Evaporation of the chloroform fraction afforded an oil which was
purified by distillation at 150 ꢀC in vacuo.Yield: 0.920 g (55%). Anal.
Calcd (%) for C6H17N3 0.15H2O (MW = 133.92 g molꢀ1): C, 53.81;
6.75ꢀ6.85 (m, 2H, HγPhOH), 6.85ꢀ6.90 (m, 2H, HRPhOH), 7.0ꢀ7.1
0
0
(m, 2H, Hβ PhOH), 7.06 (s, 2H, Hortho ,arom), 7.15ꢀ7.25 (m, 2H,
HβPhOH), 7.35ꢀ7.45 (m, 2H, Hmeta,arom), 7.8ꢀ7.9 (m, 4H, Hortho,arom),
8.04 (s, 4H, CHdN), 10.22 (s, br, PhOH). 13C NMR (100 MHz,
CDCl3) δ (ppm): 55.6 (CHdNꢀCH2ꢀCH2ꢀN), 58.9 (NꢀCH2ꢀ
PhOH), 59.6 (CHdNꢀCH2ꢀCH2ꢀN), 116.6 (CRPhOH), 119.1
(CγPhOH), 123.0 (CqRPhOH), 128.6, 128.7, 128.9, 129.0 (Cortho, Cmeta,
0
0
CβPhOH, Cβ PhOH), 130.5 (Cortho ), 136.2 (Cqarom), 157.8 (CꢀOH),
161.7 (CHdN). FT-IR ν (cmꢀ1): 3185 (OH), 2837, 2805 (CꢀH),
1642 (CdN), 799 (CꢀH ar), 746 (CꢀH PhOH), 691 (CꢀH ar).
HRMS (m/z): [M + H]+, 615.3451 (100%); calcd mass, 615.3448.
EtN(CH2CN)2. In a round-bottom flask containing 70% ethylamine
(2 mL, 25 mmol), water (15 mL), and HCl (6 mL), a 4.2 mL sample of
37% formaldehyde (55 mmol) was added, and the mixture was stirred
for 30 min. The solution was then cooled to 0 ꢀC, and NaCN (2.94 g, 55
mmol) was added. The mixture was allowed to react at room tempera-
ture for 24 h. Then, NaOH (1 g) and dichloromethane (15 mL) were
added, the organic phase was extracted, and the aqueous phase was
washed with dichloromethane (2 ꢁ 15 mL). The combined organic
fractions were dried over MgSO4, and the solvent was removed in the
rotary evaporator. The oil obtained was then purified via flash chroma-
tography in silica gel, using a hexane/ethyl acetate mixture (2:1) as
3
3
H, 13.02; N, 31.38. Found: C, 53.64; H, 13.53; N, 31.59. 1H NMR (400
MHz, CDCl3) δ (ppm): 1.01 (t, J = 7 Hz, 3H, NꢀCH2ꢀCH3), 1.31 (s,
br, 4H, NꢀCH2ꢀCH2ꢀNH2), 2.40ꢀ2.60 (m, 6H, NꢀCH2ꢀCH3 +
NꢀCH2ꢀCH2ꢀNH2), 2.75 (t, J = 7 Hz, 4H, NꢀCHꢀCH2ꢀNH2).
13C NMR (400 MHz, CDCl3) δ (ppm): 11.77 (NꢀH2ꢀCH3), 39.86
(NꢀCH2ꢀCH2ꢀNH2), 47.74 (NꢀCH2ꢀCH3), 56.75 (NꢀCH2ꢀ
CH2ꢀNH2). FT-IR ν (cmꢀ1): 3354, 3289 (NH2), 2962, 2934, 2870,
2804 (CꢀH), 1460 (ꢀCH2ꢀ, CH3), 918, 864 (NH2).
bsm2Et (L3). The procedure is the same as that for bsm2py, starting
with H2NꢀC2NEt (0.300 g, 2.29 mmol) in MeCN (40 mL) and
isophthalaldehyde (0.308 g, 2.29 mmol) in MeCN (40 mL). The
product is obtained as a solid. Yield: 0.250 g (48%). Anal. Calcd (%)
for C28H38N6 (MW = 458.64 g molꢀ1): C, 73.33; H, 8.35; N, 18.32.
3
1
eluent. Yield: 1.274 g (42%). Anal. Calcd (%) for C6H9N3 0.25H2O
Found: C, 73.29; H, 8.34; N, 18.24. H NMR (400 MHz, CDCl3) δ
3
(MW = 127.66 g molꢀ1): C, 56.45; H, 7.50; N, 32.92. Found: C, 56.37;
(ppm): 0.96 (t, J = 7 Hz, 6H, NꢀCH2ꢀCH3), 2.55 (q, J = 7 Hz, 4H,
3
H, 7.42; N, 32.71. 1H NMR (400 MHz, CDCl3) δ (ppm): 1.17 (t, J = 7
Hz, 3H, NꢀCH2ꢀCH3), 2.72 (q, J = 7 Hz, 2H, NꢀCH2ꢀCH3), 3.62
(s, 4H, NꢀCH2ꢀCN). 13C NMR (100 MHz, CDCl3) δ (ppm): 12.4
(ꢀCH3), 41.7 (ꢀCH2ꢀCH3), 48.01 (ꢀCH2ꢀCN), 114.35 (ꢀCN).
FT-IR ν (cmꢀ1): 2978, 2944, 2834 (CꢀH), 1428 (ꢀCH2ꢀ), 1106, 868.
ftN-C2NEt. A mixture of ftNC2H (5.000 g, 13.76 mmol), K2CO3
(2.850 g, 20.64 mmol), and iodoethane (2.2 mL, 27.52 mmol) in 150 mL
of acetonitrile was refluxed for 18 h. After the reaction mixture was
cooled to room temperature, it was filtered, and the solvent was
evaporated to dryness. The residue was redissolved in 100 mL of CHCl3
and was washed with 3 N aqueous NaCl solution. The aqueous phase
was extracted three times with 3 ꢁ 20 mL CHCl3. The combined
organic phases were dried over Na2SO4. Evaporation of the solvent gave
a yellow oil, which turned solid under high vacuum. Yield: 5.10 g (95%).
NꢀCH2ꢀCH3), 2.85 (t, J = 6 Hz, 8H, NꢀCH2ꢀCH2ꢀNbz), 3.66
0
(t, J = 6 Hz, 8H, NꢀCH2ꢀCH2ꢀNbz), 7.07 (s, 2H, Hortho ,arom), 7.43
(t, J = 7 Hz, 2H, Hortho,arom), 7.91ꢀ7.96 (m, 4H, Hmeta,arom), 8.04 (s,
4H, CHdN). 13C NMR (400 MHz, CDCl3) δ (ppm): 12.37
(NꢀCH2ꢀCH3), 48.57 (NꢀCH2ꢀCH3), 54.35 (NꢀCH2ꢀCH2ꢀ
NdCH), 60.00 (NꢀCH2ꢀCH2ꢀNdCH), 128.32, 128.88, 129.72
0
(Cortho , Cmeta, Cortho), 136.72 (Cqarom), 161.10 (CHdN). HRMS
(m/z): [M + Na]+, 481.3051 (100%); calcd mass, 481.3056.
Synthesis of CuI Complexes. [Cu2(L1)](PF6)2, 1(PF6)2. [Cu-
(CH3CN)4]PF6 (0.373 g, 1 mmol) was added to a solution of bsm2py
(0.292 g, 0.5 mmol) in MeOH (20 mL), and the mixture was stirred for
1 h at room temperature. The resulting yellow-orange precipitate was
filtered and washed with Et2O. The solid was then redissolved in CH2Cl2
and filtered. The obtained solid was discarded, and the solvent was
removed under vacuum to obtain the product. Yield: 0.400 g (0.40
mmol; 80%). The diffusion of a mixture of THF/Et2O (1:1) into the
mother solution yielded a dark yellow powder. Anal. Calcd (%) for
C36H40Cu2F12N8P2 0.5CH2Cl2 (MW = 1044.24 g molꢀ1): C, 41.98;
Anal. Calcd (%) for C22H21N3O4 (MW = 391.42 g molꢀ1): C, 67.51;
3
H, 5.41; N, 10.74. Found: C, 67.20; H, 5.42; N, 10.85. 1H NMR (400
MHz, CDCl3) δ (ppm): 0.95 (t, J = 6 Hz, 3H, NꢀCH2ꢀCH3), 2.65 (q,
J = 6 Hz, 2H, NꢀCH2ꢀCH3), 2.80 (t, J = 6 Hz, 4H, ftNꢀCH2ꢀ
CH2ꢀN), 3.75 (t, J = 6 Hz, 4H, ftNꢀCH2ꢀCH2ꢀN), 7.67ꢀ7.78 (m,
8H, Har). 13C NMR (400 MHz, CDCl3) δ (ppm): 11.91 (NꢀCH2ꢀ
CH3), 35.92 (ftNꢀCH2ꢀCH2ꢀN), 47.24 (NꢀCH2ꢀCH3), 51.21
(ftNꢀCH2ꢀCH2ꢀN), 123.05 (Carom), 132.22 (Cqarom), 133.68
(Carom), 168.26 (C=O). HRMS (m/z): [M + Na]+, 414.1436
(100%); calcd mass, 414.1430.
3
3
H, 3.96; N, 10.73. Found: C, 41.87; H, 3.95; N, 10.77. FD-MS (70 eV,
CH3CN): m/z = 857 (71%) [Cu2L1(PF6)]+, 791 (20%) [CuL1(PF6)]+,
647 (100%) [CuL1]+, 356 (17%) [Cu2L1]2+. IR (KBr) ν (cmꢀ1): 2914,
2857 (CꢀH), 1637 δ(CdN), 1440 δ(CꢀHaliphatic), 842 (PꢀF), 764/
689 δ(CꢀHarom).
[Cu3(L4)](PF6)3, 4(PF6)3. [Cu(CH3CN)4]PF6 (0.373 g, 1 mmol) was
added to a solution of H2NC2py (0.194 g, 1 mmol) and isophthalalde-
hyde (0.134 g, 1.00 mmol) in MeOH (20 mL), and the mixture was
stirred for 2 h at room temperature. The resulting orange precipitate was
filtered off, washed with a small amount of MeOH and Et2O, and dried
under vacuum. Yield: 0.342 g (0.288 mmol; 68%). Recrystallization of
the crude product from CH3CN and diffusion of a mixture of THF/
Et2O (1:1) into the mother solution for about 2 weeks yielded red
crystals suitable for X-ray diffraction analysis. Anal. Calcd (%) for
H2N-C2NEt. The compound H2N-C2NEt was synthesized by the
following two procedures: Method A and Method B.
Method A. A round-bottom flask, kept under nitrogen, containing
LiAlH4 (3.737 g, 95 mmol) and dry THF (110 mL), was cooled to
ꢀ10 ꢀC. Concentrated H2SO4 (5 mL) was carefully added, the mixture
was stirred for 30 min at ꢀ10 ꢀC, and then, it was allowed to warm to
room temperature. EtN(CH2CN)2 (1.274 g, 10.4 mmol) was dissolved
in dry THF (10 mL), added carefully to the hydride mixture, and allowed
to react overnight. Then, water (7 mL) was added slowly, the mixture
was stirred for 24 h, and the solvent was evaporated through a N2 stream.
Afterward, dichloromethane (50 mL) and methanol (50 mL) were
added, and the mixture was stirred again for 24 h. The solid obtained was
then filtered off and discarded, and the filtrate was evaporated. The
product was finally obtained through distillation under reduced pres-
sure. Yield: 0.291 g (21%).
C54H60Cu3N12F18P3 (MW = 1502.67 g molꢀ1): C, 43.16; H, 4.02;
3
N, 11.19. Found: C, 43.34; H, 4.30; N, 11.06. IR (KBr) ν (cmꢀ1): 2916,
2854 (CꢀH), 1633 (CdN), 1439 (CꢀH), 842 (PꢀF), 763/766
δ(CꢀH). FD-MS (70 eV, CH3CN): m/z = 939 (95%) [CuL4]+, 876
(100%) [L4 + H]+.
[Cu3(L4)](SbF6)3, 4(SbF6)3. A solution of [Cu(CH3CN)4]SbF6 (0.048
g, 0.10 mmol) in CH3CN (0.5 mL) was added dropwise to a suspension
6881
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
Table 1. Crystallographic Data for Structures L2, 4(SbF6)3, 4(PF6)3, 2(CF3SO3)2, 5(CF3SO3)3, and 6(CF3SO3)3
4(PF6)3 2.5THF
2(CF3SO3)2
5(CF3SO3)3
6(CF3SO3)3
3
3
3
3
3
structure
L2
4(SbF6)3
0.5H2O 0.75MeOH
MeCN
2 H2O
1EtOEt
3
empirical formula C38H42N6O2 C54H59Cu3F18N12Sb3 C64.75H84Cu3F18N12O3.75P3 C48H51Cu2N9O8F6S2 C66H76Cu3F9N12O14S3 C55H76Cu3N12F9O10S3
formula weight
temperature, K
wavelength, Å
crystal system
space group
a, Å
614.52
1774.00
100(2)
0.71073
triclinic
P1
1715.97
100(2)
1163.16
100(2)
0.71073
monoclinic
P21/n
1719.19
300(2)
0.71073
triclinic
P1
1523.08
373(2)
300(2)
0.71073
monoclinic
P2(1)
0.71073
triclinic
P1
0.71073
triclinic
P1
13.357(5)
90.00
12.008(18)
109.82(3)
15.72(2)
104.97(3)
20.01(3)
97.76(3)
3328(9)
2
11.4563(3)
80.620(2)
17.6766(6)
81.518(2)
36.255(1)
82.902(2)
7127.1(4)
4
15.203(3)
90
15.72(2)
104.45(3)
16.73(2)
111.79(3)
18.04(3)
106.39(3)
3878(10)
1
11.383(6)
84.485(10)
14.732(8)
80.547(9)
20.652(11)
86.199(10)
3396(3)
2
R, deg
b, Å
35.402(13)
101.776(7)
14.629(6)
90.00
18.211(3)
99.371(3)
19.084(3)
90
β, deg
c, Å
γ, deg
vol, Å3
6772(4)
8
5213.3(16)
4
Z
F (g/cm3)
Ra [I > 2σ(I)]
wRb
1.205
1.770
1.599
1.482
1.462
1.489
0.0563
0.1057
0.0869
0.0399
0.0617
0.0818
0.0898
0.1371
0.2698
2
0.1976
4 1/2
0.2416
0.2699
a R = ∑[Fo ꢀ Fc]/∑Fo. b wR = [∑(w(Fo ꢀ Fc2)2)/∑wFo ]
.
of L1 (0.029 g, 0.05 mmol) in CH3CN (0.5 mL), and the solution was
stirred for 1 h. Slow diethyl ether diffusion into the solution for about 2
weeks afforded orange crystals, which have been characterized by X-ray
diffraction analysis. Yield: 0.042 g (69%). Anal. Calcd (%) for
characterized by X-ray diffraction analysis. Yield: 0.020 g (71%). Anal.
Calcd (%) for C45H57Cu3N9F9O12S3 2.25CH3CN 0.75C4H10O
3
3
(MW = 1473.66 g molꢀ1): C, 42.79; N, 10.69; H, 4.87; S, 6.53. Found:
3
C, 42.86; N, 10.58; H, 4.73; S, 6.38. 1H NMR (400 MHz, CD3COCD3)
δ (ppm): 1.1ꢀ1.3 (m, 3H, NꢀCH2ꢀCH3), 2.6ꢀ3.1 (m, 6H,
NterꢀCH2ꢀ), 3.6ꢀ4.0 (m, 4H, CHdNꢀCH2ꢀCH2ꢀNter), 7.7ꢀ8.8
(mm, 6H, Har + CHdN). FT-IR ν (cmꢀ1): 1631 (CdN), 1253, 1223
(CF3SO3), 1149 (def ꢀCH2ꢀ), 1027 (CF3SO3), 634 (CF3SO3).
X-ray Diffraction Studies. The complexes were crystallized as de-
scribed in the synthetic procedure. Crystals of L2, 2(CF3SO3)2,
4(SbF6)3, 5(CF3SO3)3, and 6(CF3SO3)3 were mounted on a nylon
loop and used for X-ray structure determination at room temperature. The
measurements were carried out on a Bruker Smart Apex CCD diffract-
ometer. Single crystals of 4(PF6)3 were coated with polyfluorether oil
and mounted on a glass fiber. The data were collected on a Nonius
Kappa diffractometer with a CCD array detector at 173(2) K. Mo KR
radiation was used for all measurements (λ = 0.71073 Å). Space groups
were determined from systematic absences and subsequent least-squares
refinement. The structures were solved by direct methods and refined on
F2 using full-matrix least-squares techniques.21 The non-hydrogen
atoms were refined anisotropically. The H-atoms were placed in a
geometrically optimized arrangement and treated with a riding model,
except the OꢀH hydrogen atoms for the 2(CF3SO3)3, which are refined
without constraints. For the structure 5(CF3SO3)3 a considerable
amount of electron density that is attributable to partially disordered
solvent water molecules was removed with the SQUEEZE option of
PLATON.22 Those solvent molecules are, however, included in the
reported chemical formula and derived values (e.g., formula weight,
F(000), etc). Structures 2(CF3SO3)3, 5(CF3SO3)3, and 6(CF3SO3)3
present disorder on one of the CF3SO3ꢀ counterions. For the structure
4(PF6)3, SAME-restraints where used to refine solvate molecules (THF
and MeOH). Further crystallographic experimental details are given in
Tables 1 and 2 and in the Supporting Information.
C54H60Cu3N12F18Sb3 (MW = 1775.02 g molꢀ1): C, 36.54; N, 9.47;
3
H, 3.41. Found: C, 36.86; N, 9.70; H, 3.72. 1H NMR (400 MHz,
CD3COCD3) δ (ppm): 2.8ꢀ3.3 (m, 4H, NꢀCH2ꢀCH2ꢀNd),
3.4ꢀ4.0 (m, 4H, NꢀCH2ꢀCH2ꢀNd), 4.19 (s, 2H, pyꢀCH2ꢀN),
7,2ꢀ8.8 (m, 10H, Har + CHdN). FT-IR ν (cmꢀ1): 2916, 2853 (CꢀH),
1633 (CdN), 1603 (SbF6), 1440 (def ꢀCH2ꢀ), 764 (def C-Har), 653
(SbF6).
[Cu2(L2)(CH3CN)2](CF3SO3)2 2MeCN H2O, 2(CF3SO3)2 2MeCN H2O.
3
3
3
3
A solution of [Cu(CH3CN)4][CF3SO3] (0.050 g, 0.128 mmol) in
MeCN (2 mL) was added to a suspension of L2 (0.040 g, 0.065 mmol) in
MeCN (1 mL). The yellow solution was stirred for 1 h, and then it was
filtered. Addition of diethyl ether (50 mL) into the yellow solution
generated a yellow powder. Yield: 0.060 g (80%). ESI-MS (m/z): 615.3
[L2 + H]+, 637.2 [L2 + Na]+, 677.2. [L2 + Cu]+. 1H NMR (200 MHz,
CD3COCD3) δ (ppm): 2.9ꢀ3.3 (m, 4H, dNꢀCH2ꢀCH2ꢀN), 3.9ꢀ
4.4 (m, 6H, =NꢀCH2ꢀCH2ꢀN + NꢀCH2ꢀPhOH), 6.8ꢀ7.4 (mm,
5H, PhOH), 7.8ꢀ9.2 (mm, 6H, Har + CHdN). FT-IR ν (cmꢀ1): 3320
(OH), 2916, 2855 (CꢀH), 1628 (CdN), 1275, 1221, 1025 (CF3SO3),
758 (CꢀH, PhOH), 634 (CF3SO3). Anal. Calcd (%) for C60H63Cu3N9-
F9O12S3 2CH3CN H2O (MW = 1660.1 g molꢀ1): C, 46.31; N, 9.28;
3
3
3
H, 4.31; S, 5.79. Found: C, 46.12; N, 9.31; H, 4.29; S, 5.77.
The compound was also synthesized in MeOH as described initially,
and MS analysis indicated the formation of 22+ only. However, the slow
diethyl ether diffusion into the acetonitrile solution (first synthesis) of
the compound afforded a mixture of yellow and orange crystals after
12ꢀ15 days, which were both characterized by X-ray diffraction analysis
and show the formation of [Cu2(L2)(CH3CN)2](CF3SO3)2, 2(CF3SO3)2
and [Cu3(L5)](CF3SO3)3, 5(CF3SO3)3. MS (m/z): 615 [L2 + H]+,
922 [L5 + H]+.
[Cu3(L6)(CH3CN)3](CF3SO3)3, 6(CF3SO3)3.
A solution of [Cu-
(CH3CN)4] CF3SO3 (0.025 g, 0.064 mmol) in CH3CN (2 mL) was
3
’ RESULTS AND DISCUSSION
Dinuclear CuI complexes containing a macrocyclic ligand
obtained from the condensation of isophthalaldehyde and a
added to a suspension of L3 (0.015 g, 0.032 mmol) in CH3CN (0.5 mL),
and the mixture was stirred for 1 h. Slow diethyl ether diffusion into the
solution for about 2 weeks afforded yellow crystals, which have been
6882
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
Scheme 2. Potential Condensation Products from Reaction of Isophthalaldehyde and N2-Substituted Triaminea
a Labeling note: the numerical values indicate the number of reacted units and, in parentheses, the unreacted groups. For instance [1+1](n,o) means a
condensation of one molecule of isophthalaldehyde and one of triamine. In parentheses the (n) indicates an unreacted amine and (o) indicates an
unreacted aldehyde.
diethylenetriamine (with R = H in the drawing below, abbre-
viated as mac from now on) have been described previously.23
preparation of the dicyano derivative that is then reduced to the
corresponding amine by LiAlH4.
Synthesis of Metal Free [2+2] Macrocyclic Ligands. The
direct, metal-free, reaction of a dialdehyde and a diamine can
yield a large range of condensation products both macrocyclic
and acyclic, as shown in Scheme 2, that can be in equilibrium.
The relative amount of each product depends basically on
entropic and geometric factors. From an enthalpic viewpoint, it
involves the formation and breaking of the same type of bond,
and highly strained systems will be enthalpically disfavored. The
relative formation of the products shown in Scheme 2 is also
influenced by the solvent, reaction temperature, reaction time,
and, very importantly, their solubility. This wide range of
condensation compounds has been previously described in the
literature for related systems (e. g. for the pyridyldialdehyde
system).10b,26
In our case, the [2+2] macrocyclic ligands were prepared by a
condensation of a 1:1 molar ratio of an adequately substituted
triamine and isophthalaldehyde that was very slowly added to
the respective triamine solution to favor both lower oligomeric
compounds as well as macrocyclic type products. The relatively
low yields obtained indicate the formation of other products and
potentially unreacted starting materials. Once the [2+2] con-
densation product was formed, it was redisolved in either MeOH
or MeCN and was stirred at room temperature (RT) for 24 h. MS
analysis of the solution indicates the presence of the [2+2]
condensation product only. Thus, once it is formed, and in the
absence of a catalyst, there is no equilibration process that could
generate a mixture of oligomers, at least on the time scale of days.
Synthesis of CuI Complexes. Another factor that strongly
influences reactivity is the presence of a metal cation that can act
as a templating agent and thus stabilize the formation of a
condensation product that possesses a cavity size and shape that
is complementary to those of the templating cation.
The derived CuI complex was especially interesting because it
undergoes an intramolecular ligand hydroxylation reaction when
treated with dioxygen.24 This reaction can be regarded as a model
reaction for the enzymatic reaction of tyrosinase, a monooxy-
genase that is responsible for o-hydroxylation of the phenol
entity.25
To gain more insight into the interesting properties of this
macrocyclic ligand type, we now have investigated how the
modification of the R group in [Cu2(mac)(CH3CN)2]2+ from
H to an ethyl group, 2-methylpyridyl, and 2-methylphenol would
influence the coordination behavior of these systems.
Synthesis of Macrocycle Components. For the preparation
of the substituted triamines, a multistep process, shown in the
upper part of Scheme 1, was followed, consisting of the following:
(a) the protection of the primary amines with phthalic anhydride
to form the corresponding phthalimides; (b) the addition of the
pyridyl or phenol aldehyde or iodoethane to the central amine;
(c) deprotection of the phthalamides with hydrazine to yield the
corresponding primary amines. The ethyl substituted central
amine was also prepared by following a different synthetic
strategy depicted at the bottom of Scheme 1A. It describes the
6883
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
Figure 1. Ball and stick diagrams for the X-ray crystal structure for CuI complexes: (A) top, two representations of 22+; bottom, two of 53+; (B) left, 43+;
right, 63+. Color codes: Cu, orange; N, blue; C, gray; O, red. H atoms are not shown.
The synthesis of the CuI complexes was carried out either by
mixing 2 equivalents of the [CuI(MeCN)4]+ salt and one of the
[2+2] free ligand or via a template procedure as indicated in the
following equations. For the case of the L1 ligand, the solvent and
crystallization time have a strong influence over the complexes
obtained. In MeOH, the main product obtained in 80% yield is 12+,
as indicated in eq 1.
transformation that generates L4 out of L1, which will be
discussed later. Further, a one pot synthesis using the triamine
and dialdehyde and CuI as a template metal generates the 43+
complex in 68% yield, as shown in eq 2.
+
3½CuIðMeCNÞ4ꢂ + 3½1; 3-PhðCHOÞ2ꢂ
MeOH
3+
s
½Cu3ðL4Þꢂ
ð2Þ
MeOH, RT
+
2+
+ 3H2NC2py
f
2½CuIðMeCNÞ4ꢂ + L1
½Cu2ðL1Þꢂ + 8MeCN ð1Þ
f
43+
12+
On the other hand, and in sharp contrast, using MeCN as the
solvent generates the analogous trinuclear complex 43+ in 70%
yield. Given the fact that the L1 ligand does not isomerize
in solution, it suggests the presence of a metal assisted
For the case of the L2 ligand, only the dinuclear complex, 22+,
is obtained in either MeOH or MeCN in good yields
(approximately 80%) after 1 h of mixing the reactants at room
temperature. However, if the solution is allowed to stand for
6884
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
Figure 2. Variable temperature 1H NMR spectra of 1(PF6)2 in DMFꢀ
d7 at ꢀ55, ꢀ15, and +25 ꢀC (arrows mark the peaks of the second
isomer).
12ꢀ15 days, a mixture of dinuclear, 22+, and trinuclear, 53+,
complexes is obtained (see eq 3).
2+
½Cu2ðL1ÞðMeCNÞ2ꢂ
MeCN, RT
+
½CuIðMeCNÞ4ꢂ + L2
f
22+
2+
+ ½Cu2ðL5ÞðMeCNÞ3ꢂ
ð3Þ
53+
Finally, for the case of the L3 macrocyclic ligand, only the
trinuclear complex was obtained in 71% yield, indicating the
formation of the L6 ligand, as shown in eq 4.
MeCN, RT
+
3+
6½CuIðMeCNÞ4ꢂ + 3L3
2½Cu3ðL6ÞðMeCNÞ3ꢂ
f
63+
ð4Þ
Solid State Characterization. The crystal structure of the
ligand L2 consists of eight discrete L2 molecules (see Table 1).
The X-ray analysis shows four crystallographically independent
but chemically identical L2 units, which present very slight
variations in bond distances and angles. It is interesting to note
that, for each molecular structure, the two benzene rings are
nearly parallel to one another with an angle of 1.76, 1.84, 6.27, or
6.96ꢀ and that the phenol groups are placed in mutually trans
position in an inversion center arrangement around the tertiary
amine, permitting the establishment of H-bonding with the
nearby units. In Figure 1 ball and stick representations of the
X-ray structures for the dinuclear complex, 22+, and the trinuclear
complexes, 43+, 53+, and 63+, is depicted. For the dinuclear
complex, 22+, the metasubstitution of the aromatic ring places the
two copper centers at a distance of 7.97 Å; whereas the two
benzene rings are nearly parallel to one another with an angle of
29.1ꢀ. Each copper center has a distorted tetrahedral arrange-
ment as a result of the constraints imposed by the triaza moiety of
the macrocyclic ligand. This generates a long CuꢀN (2.208 Å)
distance with the central amine group, two medium CuꢀN
(2.022 Å and 2.061 Å) distances with the imines, and a short
distance with the MeCN monodentate ligand CuꢀN (1.918 Å).
The strain of the macrocyclic ligand also imposes two short
NꢀCuꢀN angles of 84.85ꢀ and 83.95ꢀ, with the rest of the
6885
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
Figure 3. ESI-MS spectra obtained for 2(CF3SO3)2 (top) and 5(CF3SO3)3 (bottom) in MeCN.
NꢀCuꢀN angles ranging 111ꢀ128ꢀ. Finally, the dangling
phenol group is not coordinating the Cu metal center. The
metric parameters described here are also in agreement with
related CuI complexes that have been previously reported in the
literature.27 Table 2 lists selected bond distances and angles for
the first coordination sphere of one of the CuI metal centers of
complexes 4, 2, 5, and 6.
are practically identical. For the trinuclear complex 43+, the
dangling pyridyl group is coordinating the metal center, replacing
the MeCN when compared to 53+. For the trinuclear complexes,
it is also interesting to see that the three benzene rings altogether
adopt a bowl shape arrangement. For the case of 43+, the closest
H atoms among the three aromatic rings are situated at 2.52, 2.70,
and 2.77 Å, forming an irregular triangle, and the angles between
these aromatic rings are 56.7, 69.0, and 83.2ꢀ. Withregardto the3D
packing of these molecules, it is interesting to realize that those
containing the triflate anion have packing that is dominated by
H-bonding with triflate oxygen atoms and the macrocycle. A similar
situation is found for complex 43+ containing PF6ꢀ as counteranion
that crystallizes with THF, H2O, and MeOH. Here again, packing
interactions are dominated by extensive hydrogen bonding be-
tween the solvate oxygen atoms and the macrocycle. However,
complex 43+ containing SbF6ꢀ as counteranion crystallizes with no
solvate molecules, and its packing structure is significantly different
For the trinuclear complexes 53+ and 63+, the local CuI
coordination is comparable to that of the dinuclear 22+ complex.
Here, the metal centers are disposed in a triangular arrangement
with CuꢀCu distances (ranging from 8.8 to 9.4 Å) that are a bit
larger than those for the dinuclear complex, as discussed above,
and thus manifest the relative flexibility of this family of [2+2]
and [3+3] Schiff base ligands. Comparing the 53+ trinuclear
complex and the 22+ dinuclear complex, the major difference is
that the latter has slightly shorter NimꢀCu(1)ꢀNim angles,
119ꢀ123ꢀ vs. 108ꢀ111.32ꢀ, while the Cu bonding distances
6886
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
from the rest. In particular, it is interesting to see the presence of
dimers of trinuclear units bonded by πꢀπ and CHꢀπ interactions
between macrocyclic ligands (see the Supporting Information.).
1
NMR Spectroscopy. The H NMR spectrum of 12+ was
recorded in CD3CNꢀd3 or in DMFꢀd7 and is presented in
Figure 2. At room temperature, the spectra show very broad
signals that indicate the presence of a dynamic behavior analo-
gous to that previously described for [Cu2(mac)(CH3CN)2]2+.24b
Whereas the aliphatic part is unremarkable, the aromatic part
displays sharp resonances at ꢀ25 ꢀC and below. Together
with these sharp resonances, lower intensity and wider peaks
also appear that are presumably the result of another highly
symmetric stereoisomer of 12+, given the reduced number of
resonances observed. It is also interesting to observe that
resonance for Ha in 12+ appears at 10.22 ppm, while for
[Cu2(mac)(CH3CN)2]2+ it is shifted to 8.74 ppm, manifesting
how subtle differences in structure can produce large electronic
perturbation at a specific site.
Figure 4. Graph of the 22+ to 53+ oligomerization evolution as a
function of time monitored by using MS spectroscopy. The inset shows
the first 30 h. Units are the same as in the main graph.
The CuI complexes described here, in general, react slowly
(within the time scale of days) with molecular oxygen, except
for the ones containing the phenolic ligand, 22+ and 53+, which
react a bit faster, presumably to form the corresponding bis-μ-
hydroxo derivatives Cu(μꢀOH)2Cu, as has been previously
shown for related metasubstituted macrocyclic complexes.11
Unlike [Cu2(mac)(CH3CN)2]2+, no hydroxylation of the
ligand occurs with the complexes described in the present
work.
MS Spectroscopy and the [2+2] vs. [3+3] Evolution Pro-
cess. Complexes 22+ and 53+ were analyzed by ESI-MS, and their
spectra are presented in Figure 3. In both cases, their molecular
peaks could not be identified, but a series of fragments are found.
For complex 22+, key monocharged peaks at 615 (L2 + H, highest
intensity), 637 (L2 + Na), 677 (L2 + Cu), 739 (L2 + Cu2 ꢀ 1),
and 766 m/z (L2 + Cu2Na ꢀ 2) could be identified. For complex
53+, key peaks are found at 922 (L5 + H, highest intensity), 984
m/z (L5 + Cu), and their corresponding doubly charged peaks at
461 and 492 m/z, respectively. For both complexes, the relative
intensities of their peaks coincide perfectly with the simulated
ones. As indicated in the previous section, the reaction of
[CuI(MeCN)4]+ in MeCN with the [2+2] condensation macro-
cyclic ligand L2 generates a mixture of the dinuclear and tri-
nuclear complexes as indicated in eq 3. Thus, it can be inferred
that an equilibrium between the dinuclear and trinuclear complex
may exist, as indicated in the following equation:
extrapolated at a synthetic level with regard to the relative
amount of 22+ and 53+ because in that case we used a mixture
of MeCN and ether.
The formation of the trinuclear complex from the dinuclear
compound indicates that at least one of the imine CdN of the
[2+2] Schiff base ligand has to be broken, and then the fragments
have to react again so that the new [3+3] ligand can be formed.
This process has not been observed for the free ligand, at least
during the time scale of days. Thus, it must be assisted by the CuI
dinuclear complex. This is in sharp contrast with the cases of
other macrocyclic ligands where this process is known to occur
very quickly, as is the case for the systems derived from pyridine
dialdehyde and diamine.26 Scheme 2 presents potential con-
densation products that can be obtained from the reaction of a
1:1 dialdehyde and triamine to illustrate the variety of com-
pounds, including macrocyclic and acyclic compounds. As
mentioned earlier, the L2 does not undergo any reorganization
process by itself, but it does so when complexed to CuI ions.
Thus, potential fragments that can lead to the trinuclear com-
plex are as follows:
2+
+
3+
fCu2½2 + 1ðo2Þꢂg + fCu½1 + 2ðn2Þꢂg f ½Cu3ðL5Þꢂ
53+
ð6Þ
or
3+
2+
2½Cu3ðL5ÞðMeCNÞ3ꢂ
MheCN
ð5Þ
3½Cu2ðL2ÞðMeCNÞ2ꢂ
22+
53+
+
3+
fCu2½2 + 2ðn, oÞꢂg2+ + fCu½1 + 1ðn, oÞꢂg f ½Cu3ðL5Þꢂ
53+
This reaction was monitored using MS, following the relative
intensities of the peaks at 615 m/z for 22+ and 922 m/z for 53+, at
room temperature. The initial concentration of complex 22+ was
0.026 M, and no 53+ was observed. As time elapsed, the
formation of 53+ was observed, as depicted in Figure 4. After
1.5 months, the system reaches equilibrium with a relative
concentration [53+]/[22+] of 0.65, which implies an equilibrium
constant of 0.42 for eq 5. This value indicates that the [2+2]
condensation complex 22+ is more energetically favored than the
[3+3] complex 53+, probably as a result of entropic factors and
also, to a minor extent, of the relative strain of their structures. It
is important to bear in mind that these experiments have been
carried out under high dilution conditions so that both com-
plexes are completely soluble. Therefore, these results cannot be
ð7Þ
The ligand nomenclature is described in Scheme 2. For
instance, for the case of [2 + 1(o2)], the [2+1] indicates the
condensation product of two dialdehydes and one triamine and
in the parentheses is indicated the number and nature of
unreacted groups, n for a secondary amine and o for aldedyde.
Final Comments and Conclusions. In this paper, we report
the synthesis of three new [2+2] macrocyclic ligands, L1, L2, and
L3, with pendant arms that consist of the condensation of
phthalaldehyde and N2-substituted triamines with moderate to
good yields. We have also shown that once these [2+2] com-
pounds are formed, they do not undergo further rearrangement
reactions in solution, and thus, the different compounds shown
6887
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
in Scheme 2 are not in equilibrium in our case. Therefore, the
formation of the [1+1] condensation product, higher oligomers,
such as [3+3], [4+4], etc., and linear polymers represents very
minor products, and thus the [2+2] condensation product is the
favored one.
(11) Costas, M.; Ribas, X.; Poater, A.; Lꢁopez Valbuena, J. M.; Xifra,
R.; Company, A.; Duran, M.; Sola, M.; Llobet, A.; Corbella, M.; Uson,
M. A.; Mahia, J.; Solans, X.; Shan, X.; Benet-Buchholz, J. Inorg. Chem.
2006, 45, 3569–3581.
(12) (a) Brooker, S.; McKee, V.; Shepard, W. B.; Pannell, L. K.
J. Chem. Soc., Dalton Trans. 1987, 11, 2555–2562. (b) Fenton, D. E.;
Kitchen, S. J.; Spencer, C. M.; Tamburini, S.; Vigato, P. A. J. Chem. Soc.,
Dalton Trans. 1988, 685–690. (c) Aspinall, H. C.; Moore, S. R.; Smith,
A. K. J. Chem. Soc., Dalton Trans. 1993, 709–714. (d) Brooker, S.; Kelly,
R. J.; Sheldrick, G. M. J. Chem. Soc., Chem. Commun. 1994, 487–488.
(e) Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; Benelli, C. Inorg.
Chem. 1995, 34, 5507–5515. (f) Brooker, S.; McKee, V.; Metcalfe, T.
Inorg. Chim. Acta 1996, 246, 171–179. (g) Givaja, G.; Blake, A. J.;
Wilson, C.; Schr€oder, M.; Love, J. B. Chem. Commun. 2005,
4423–4425. (h) Ma, C. T. L.; MacLachlan, M. J. Angew. Chem., Int.
Ed. 2005, 44, 4178–4182. (i) Radecka-Paryzek, W.; Patroniak, V.;
Lisowski, J. Coord. Chem. Rev. 2005, 249, 2156–2175. (j) Gallant, A. J.;
Hui, J. K.-H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org.
Chem. 2005, 70, 7936–7946. (k) Nabeshima, T.; Miyazaki, H.; Iwasaki,
A.; Akine, S.; Saiki, T.; Ikeda, C. Tetrahedron 2007, 63, 3328–3333.
(l) Ikeda, C.; Sakamoto, N.; Nabeshima, T. Org. Lett. 2008, 10,
4601–4604. (m) Akine, S.; Nabeshima, T. Dalton Trans. 2009, 47,
10395–10408.
(13) (a) Wenzel, M.; Bruere, S. R.; Knapp, Q. W.; Tasker, P. A.;
Plieger, P. G. Dalton Trans. 2010, 39, 2936–2941. (b) Anda, C.; Llobet,
A.; Salvado, V.; Motekaitis, R.; Riebenspies, J.; Martell, A. E. Inorg. Chem.
2000, 39, 2986–2999. (c) Anda, C.; Llobet, A.; Salvado, V.; Motekaitis,
R.; Martell, A. E. Inorg. Chem. 2000, 39, 3000–3008. (d) Anda, C.;
Llobet, A.; Martell, A. E.; Donnadieu, B.; Parella, T. Inorg. Chem. 2003,
42, 8545–8550. (e) Anda, C.; Llobet, A.; Martell, A. E.; Riebenspies, J.;
Berni, E.; Solans, X. Inorg. Chem. 2004, 43, 2793–2802. (f) Arbuse, A.;
Anda, C.; Martínez, M. A.; Perez-Miron, J.; Jaime, C.; Parella, T.; Llobet,
A. Inorg. Chem. 2007, 46 (25), 10632–10638.
In contrast, the reaction of the [2+2] condensation ligands, L1
and L2, with CuI complexes generates a mixture of dinuclear (12+
and 22+) and trinuclear (43+ and 53+) complexes that are in
equilibrium in solution. The unique reactivity of the present CuI
complex puts forward the delicate balance between electronic
and geometrical factors that allow the making and breaking of
imine bonds, and thus observation of the [2+2] and [3+3]
equilibrium reaction. Finally, all of the CuI complexes described
here react only very slowly with molecular oxygen at room
temperature and thus manifest the capacity of the Schiff base
ligand to stabilize the CuI oxidation state in a MeCN solution.
’ ASSOCIATED CONTENT
S
Supporting Information. Crystallographic information
b
files (CIF) for L2, 2(CF3SO3)2, 4(SbF6)3, 4(PF6)3, 5(CF3SO3)3,
and 6(CF3SO3)3 and additional experimental and spectroscopic
data. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Authors
*E-mail: allobet@iciq.es; angeles.martinez@udg.edu; siegfried.
schindler@chemie.uni-giessen.de.
(14) Arbuse, A.; Font, M.; Martínez, M. A.; Fontrodona, X.; Prieto,
M. J.; Moreno, V.; Sala, X.; Llobet, A. Inorg. Chem. 2009, 48, 11098–11107.
(15) (a) Poater, A.; Ribas, X.; Cavallo, L.; Llobet, A.; Sola, M. J. Am.
Chem. Soc. 2008, 130, 17710–1771. (b) Llobet, A.; Martell, A. E.;
Martinez, M. A. J. Mol. Catal. 1998, 129, 19–26.
(16) Company, A.; Jee, J. E.; Ribas, X.; Lopez-Valbuena, J. M.;
Gomez, L.; Corbella, M.; Llobet, A.; Mahia, J.; Benet-Buchholz, J.;
Costas, M.; van Eldik, R. Inorg. Chem. 2007, 46, 9098–9110.
(17) Weitzer, M.; Brooker, S. Dalton Trans. 2005, 2448.
(18) Dioury, F.; Sylvestre, I.; Siaugue, J.-M.; Wintgens, V.; Ferroud, F.;
Favre-Reguillon, A.; Foos, J.; Guy, A. Eur. J. Org. Chem. 2004, 4424–4436.
(19) Utz, D.; Kisslinger, S.; Hampel, F.; Schindler, S. J. Inorg.
Biochem. 2008, 102, 1236–1245.
(20) Song, B.; Reuber, J.; Ochs, C.; Hahn, F. E.; L€ugger, T.; Orvig, C.
Inorg. Chem. 2001, 40, 1527–1535.
’ ACKNOWLEDGMENT
This research has been financed by the MICINN of Spain
through Grant Nos. CTQ2007-67918, CTQ2010-21497, and
Consolider Ingenio 2010 CSD2006-0003. A.A. is grateful for the
award of a doctoral grant from CIRIT Generalitat de Catalunya,
Spain. D.U. acknowledges a stipend from the University of
Erlangen-N€urnberg. Furthermore, she also thanks Geoffrey
Lawrence (University of Newcastle, Australia) for his support
of the parts of this work that she performed in his laboratories.
For the stay in Australia, she acknowledges a scholarship from
the DAAD.
(21) Sheldrick, G. M. SHELXTL, Version 6.14, Program for Crystal
Structure Refinement; Bruker Advanced X-ray Solutions, Universit€at
G€ottingen: Germany, 1997, 2000ꢀ2003.
’ REFERENCES
(1) Vigato, P. A.; Tamburini, S.; Bertolo, L. Coord. Chem. Rev. 2007,
251, 1311–1492.
(22) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;
Utrecht University: Utrecht, The Netherlands, 2005.
(23) (a) Ngwenya, M. P.; Martell, A. E.; Reibenspies, J. H. Chem.
Commun. 1990, 1207. (b) Ngwenya, M. P.; Chen, D.; Martell, A. E.;
Reibenspies, J. H. Inorg. Chem. 1991, 30, 2732–2736.
(24) (a) Becker, M.; Schindler, S.; van Eldik, R. Inorg. Chem. 1994,
33, 5370–5371. (b) Ma, H.; Allmendinger, M.; Thewalt, U.; Lentz, A.;
Klinga, M.; Rieger, B. Eur. J. Inorg. Chem. 2002, 2857–2867. (c) Menif,
R. E.; Martell, A. E.; Squattrito, P. J.; Clearfield, A. Inorg. Chem. 1990,
29, 4723–4729. (d) Utz, D.; Heinemann, F. W.; Hampel, F.; Richens,
D. T.; Schindler, S. Inorg. Chem. 2003, 42, 1430–1436.
(25) (a) Claus, H.; Decker, H. Syst. Appl. Microbiol. 2006, 29, 3–14.
(b) Matoba, Y.; Kumagai, T.; Yamamoto, A.; Yoshitsu, H.; Sugiyama, M.
J. Biol. Chem. 2006, 281, 8981–8990. (c) Decker, H.; Dillinger, R.;
Tuczek, F. Angew. Chem., Int. Ed. 2000, 39, 1591–1595. (d) Fusi, V.;
Llobet, A.; Mahía, J.; Micheloni, M.; Paoli, P.; Ribas, X.; Rossi, P. Eur. J.
Inorg. Chem. 2002, 4, 987–990.
(2) Alexander, V. Chem. Rev. 1995, 95, 273.
(3) Fenton, D. E.; Okawa, H. Chem. Ber. 1997, 130, 433.
(4) Collinson, R.; Fenton, D. E. Coord. Chem. Rev. 1996, 148, 19.
(5) Fenton, D. E. Pure Appl. Chem. 1986, 58, 1437.
(6) Draho, B. S.; Kotek, J.; Hermann, P.; Luke, I.; Toth, E. Inorg.
Chem. 2010, 49, 3224–3238.
(7) Caltagirone, C.; Gale, P. A. Chem. Soc. Rev. 2009, 38, 520–563.
(8) Martell, A. E.; Penitka, J.; Kong, D. Coord. Chem. Rev. 2001,
216ꢀ217, 55–63.
(9) Nelson, J.; McKee, V.; Morgan, G. In Progress in Inorganic
Chemistry; Karlin, K. D. Ed.; John Wiley & Sons: New York, 1998;
Vol. 47, p 167.
(10) (a) Collinson, S. R.; Fenton, D. E. Coord. Chem. Rev. 1996,
148, 19–40. (b) Nelson, S. M. Pure Appl. Chem. 1980, 52, 461–2476.
(c) Brooker, S. Coord. Chem. Rev. 2001, 222, 33–56.
6888
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889

Inorganic Chemistry
ARTICLE
(26) (a) Vigato, P. A.; Tamburini, S. Coord. Chem. Rev. 2004,
248, 1717–2128. (b) Hutin, M.; Bernardinelli, G.; Nitschke, J. R. Proc.
Natl. Acad. Sci. 2006, 103, 17655–17660.
(27) (a) Costas, M.; Sola, M.; Robles, J.; Xifra, R.; Llobet, A.; Parella,
T.; Stoeckli-Evans, H.; Neuburger, M. Inorg. Chem. 2003, 42, 4456–
4468. (b) Veauthier, J. M.; Tomat, E.; Lynch, V. M.; Sessler, J. L.;
Mirsaidov, U.; Markert, J. T. Inorg. Chem. 2005, 44, 6736–6743.
6889
dx.doi.org/10.1021/ic102185y |Inorg. Chem. 2011, 50, 6878–6889