A novel di-compartmental bis-(2-hydroxyisophtalamide) macrocyclic ligand and its mononuclear Cu(ii) and Ni(ii) complexes

Article abstract of DOI:10.1039/c2dt31586f

The synthesis and characterisation of the new di-compartmental bis-(2-hydroxyisophtalamide) macrocyclic pro-ligand, LH₆, which comprises two phenol-diamide units linked by ethylene bridges, is herein reported, together with its corresponding di-phenolate salt, [NBu ₄]₂[LH₄]. The three macrocyclic compounds, [LH₄(OMe)₂] (protected ligand), LH₆ and [LH₄][NBu₄]₂ were fully characterised including X-ray crystallography for [LH₄(OMe)₂] and [NBu ₄]₂[LH₄]. The results of solid-state and solution studies have indicated that the macrocycle can adopt specific conformations, which are influenced by H-bonding interactions as well as the deviation of the amide carbonyl relative to the phenol plane. LH₆ reacts with M^II(acetate)₂·(H₂O) ₆ (M = Ni, Cu) in a 1:1 ratio in the presence of 4 eq of [NBu ₄](OH) in methanol to afford the dianionic [M(LH₂)] ^2- complexes, 1^2- and 2^2-, respectively. The X-ray crystallography, EPR, NMR and UV-vis spectroscopic data, combined with DFT calculations, indicate that 1^2- and 2^2- are unique unsymmetrical square planar mononuclear complexes that are intramolecularly H-bonded. Thus, one macrocyclic compartment contains a M^II-N ₂O₂ centre resulting from the tetra-anionic di-phenolato di-amidato ligation; the other compartment possesses two protonated amide N-H groups that are H-bonded the coordinated phenolate O atoms. This represents a unique example in which a phenolate is both coordinated and intramolecularly H-bonded. This H-bonding appears unusually strong as revealed by N(H/D) exchange experiments; and may be responsible for the stability of the mononuclear complex, and the difficulty in isolating the corresponding dinuclear complex [M₂(L)]^2-.

Full text of DOI:10.1039/c2dt31586f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 12457
www.rsc.org/dalton
PAPER
A novel di-compartmental bis-(2-hydroxyisophtalamide) macrocyclic ligand
and its mononuclear Cu(II) and Ni(II) complexes†
Meital Eckshtain-Levi,a Ronit Lavi,a Dmitry Yufit,b Maylis Orio,c Riccardo Wanked and Laurent Benisvy*a
Received 17th July 2012, Accepted 8th August 2012
DOI: 10.1039/c2dt31586f
The synthesis and characterisation of the new di-compartmental bis-(2-hydroxyisophtalamide)
macrocyclic pro-ligand, LH6, which comprises two phenol-diamide units linked by ethylene bridges, is
herein reported, together with its corresponding di-phenolate salt, [NBu4]2[LH4]. The three macrocyclic
compounds, [LH4(OMe)2] (protected ligand), LH6 and [LH4][NBu4]2 were fully characterised including
X-ray crystallography for [LH4(OMe)2] and [NBu4]2[LH4]. The results of solid-state and solution studies
have indicated that the macrocycle can adopt specific conformations, which are influenced by H-bonding
interactions as well as the deviation of the amide carbonyl relative to the phenol plane. LH6 reacts with
MII(acetate)2·(H2O)6 (M = Ni, Cu) in a 1 : 1 ratio in the presence of 4 eq of [NBu4](OH) in methanol to
afford the dianionic [M(LH2)]2− complexes, 12− and 22−, respectively. The X-ray crystallography, EPR,
NMR and UV-vis spectroscopic data, combined with DFT calculations, indicate that 12− and 22− are
unique unsymmetrical square planar mononuclear complexes that are intramolecularly H-bonded. Thus,
one macrocyclic compartment contains a MII–N2O2 centre resulting from the tetra-anionic di-phenolato
di-amidato ligation; the other compartment possesses two protonated amide N–H groups that are
H-bonded the coordinated phenolate O atoms. This represents a unique example in which a phenolate is
both coordinated and intramolecularly H-bonded. This H-bonding appears unusually strong as revealed
by N(H/D) exchange experiments; and may be responsible for the stability of the mononuclear complex,
and the difficulty in isolating the corresponding dinuclear complex [M2(L)]2−
.
extensively developed, taking advantage of so-called template
synthesis, to provide homo- and hetero-dinuclear complexes of
many transitions metal ions.4 Both salen and di-compartmental
salen are dianionic ligands, often yielding neutral or cationic
Introduction
Phenol-containing macrocyclic and open-macrocyclic ligands
such as salen,1 calixarene,2 2-hydroxyisophtalimides cages3 and
others have been extensively used to provide specific coordina-
tive environments for various transition metals, and find fascinat-
ing developments in areas as diverse as catalysis, molecular
magnetism, sensors, and medicinal chemistry.1–3 The well-
known salen (Scheme 1) open-macrocycle provides a N2O2
coordination sphere through O-phenolate and N-imine ligations
to the metal ions, stabilising thier mononuclear complexes which
have been explored in various areas.1 Di-compartmental macro-
cyclic phenol–Schiff bases (Scheme 1) providing two N2O2
chelating sites with bridging phenolate groups, have also been
aDepartment of Chemistry, Bar-Ilan University, Ramat Gan 52 900,
Israel. E-mail: benisvl@mail.biu.ac.il; Fax: +972-3-7384053;
Tel: +972-3-7384599
bDepartment of Chemistry, University of Durham, South Road, Durham
DH1 3LE, UK
cLaboratoire de Spectrochimie Infrarouge et Raman, CNRS LASIR-UMR
8516, Bat C4 F-59655 Villeneuve d’Ascq Cedex, France
dCentro de Química Estrutural, Complexo I, I.S.T, TU, Lisbon, Av.
Rovisco Pais, 1049-001 Lisbon, Portugal
†Electronic supplementary information (ESI) available. CCDC
883918–883921. For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/c2dt31586f
Scheme 1
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12457–12467 | 12457

View Article Online
complexes e.g. with divalent Cu or Ni ions. In contrast, bis-sali-
cylamide pro-ligands (Scheme 1), which possess N–H amide
groups (instead of N-imine in salen), when deprotonated,
provide a tetra-anionic N2O2 coordination sphere, often leading
to polyanionic complexes;5–7 e.g. square planar dianionic Cu(II)
and Ni(II) complexes.6 The strong N−-amidate σ-donation, the
polyanionic nature of the ligand and the presence of a redox
non-innocent phenol group, confer complexes of bis-salicyl-
amide ligands attracting properties such as their ability to stabilise
higher oxidation states of the metal ion.5 Moreover, they have
shown a strong ability to act as “complexes ligand” towards
cationic entities6,7 for the constructions of 3d/3d8 and 3d/4f9
assemblies as magnetic materials. Despite these advantages,
bis-salicylamide ligands have been largely underexploited as
compared to salen ligands, while the “all-amide version” of di-
compartmental bis-salen, i.e., bis-2-hydroxyisophtalamide
(Scheme 1), has never been reported. Only mixed salicylamide/
salicylimine di-compartmental macrocyclic ligands and their
complexes have been reported, though they lack structural
characterisation.10 We herein report the synthesis and characteris-
ation of the new bis-2-hydroxyisophtalamide macrocyclic pro-
ligand, LH6, together with its corresponding di-phenolate salt,
[NBu4]2[LH4], as well as the synthesis and characterisation of
their corresponding unsymmetrical dianionic mononuclear Ni(II)
and Cu(II) complexes, [NBu4]2[1] and [NBu4]2[2] respectively
(Scheme 2).
for 12 h at room temperature, filtered and the filtrate was evapo-
rated to yield a yellow oil in 70% yield (0.53 g). 1H NMR
(300 MHz, CDCl3 + D2O, 298 K): δ/ppm 8.08 (s, 2H, Ar), 3.9
(t, J3 = 6 Hz, 4H, CH2), 3.8 (s, 3H, CH3), 3.5 (t, J3 = 6 Hz, 4H,
CH2), 1.3 (s, 9H, tBu). MS ESI(+): m/z 337 ([4H]+, 100%).
[LH4(OMe)2]. A mixture of N1,N3-bis(2-aminoethyl)-5-(tert-
butyl)-2-methoxyisophthalamide (0.660 g, 1.96 mmol), P (0.876 g,
1.96 mmol) and Et3N (0.800 g, 7.85 mmol) in 1.8 L of CH2Cl2
was stirred at room temperature under N2 atmosphere for 72 h,
after which time the solvent was evaporated to give a yellow
solid. The latter was dissolved in CH2Cl2 and extracted with
water. The organic phase was dried over MgSO4, filtered and the
solution was evaporated to give a white solid. The crude was
then purified by silica gel flash chromatography (eluent: ethyl
acetate–methanol 90 : 10) affording [LH4(OMe)2] in 42% yield
(0.450 g). Single crystals of [LH4(OMe)2], suitable for X-ray
crystallography, have been obtained by diffusing of hexane onto
a concentrated dichloromethane solution. Elemental analysis:
Calc. for C30H40N4O6·H2O: C 63.14, H 7.42, N 9.82; Found:
C 63.52, H 7.70, N 10.17. 1H NMR (300 MHz, DMSO-d6,
298 K): δ/ppm 8.11 (br s, 4H, NH), 7.72 (s, 4H, Ar H), 3.57
t
(6H, CH3), 3.51 (br s, 8H, CH2), 1.28 (s, 18H, Bu). 13C NMR
(75 MHz, DMSO-d6, 298 K): δ/ppm 165.25 (C(O)NH), 146.7
(C–OH, Ar), 129.39 (Ar–C), 127.64 (Ar–Cmeta), 63.9 (CH3–O),
38.14 (CH2–NH), 30.95 (CH3, tBu). MS ESI(+): m/z 553
([M − H]+, 100%). UV-vis (DMSO) λmax/nm (ε/M−1 cm−1):
284 (3209).
Experimental
LH6. A suspension of [LH4(OMe)2] (0.450 g, 0.815 mmol) in
33% HBr–AcOH (12 mL) was heated to reflux (100 °C) with
vigorous stirring for one hour, during which time a precipitate
formed. The mixture was then cooled to room temperature and
water was added. The solid was then collected by filtration and
dried under vacuum to give LH6 as a white powder in 98% yield
(0.420 g). Elemental analysis: Calc. for C28H36N4O6: C 64.10,
Dry distilled CH2Cl2, CH3CN and CH3OH were purchased from
Aldrich Chemical Ltd. Commercially available, solid chemicals
were usually used without further purification. Organic chemi-
cals were purchased from Fluka, Acros and Aldrich. Ni-
(OAc)2·4H2O, Cu(OAc)2·H2O, and 5-tert-butyl-2-methoxy-
isophthalic acid was purchased from Acros organic.
1
H 6.92, N 10.68; Found: C 64.27, H 7.30, N 10.89. H NMR
(300 MHz, DMSO-d6, 298 K): δ/ppm 14.16 (s, 2H, OH), 8.77
(s, 4H, NH), 8.00 (s, 4H, Ar–H), 3.50 (s, 8H, CH2), 1.32 (s,
Ligand synthesis
t
18H, Bu). 13C NMR (75 MHz, DMSO-d6, 25 °C): δ/ppm
Bis-(2,5-dioxopyrrolidin-1-yl)-5-(-butyl)-2-methoxyisophthalate
(3). 5-(tert-Butyl)-2-methoxyisophthalic acid (2)11 (5 g, 19 mmol)
and N-hydroxysuccinimide (5.02 g, 43 mmol) were dissolved in
80 mL of dioxane under N2 atmosphere. A solution of 1,3-di-
cyclohexylcarbodiimide (9 g, 43 mmol) in 85 mL of CH2Cl2
was added dropwise to the reaction mixture at 10 °C. The result-
ing suspension was stirred at room temperature for 12 h, filtered
and the filtrate was evaporated. The solid was recrystallised
from hot ethanol to give a white powder in 47% yield (4.17 g).
168.10 (C(O)NH), 157.54 (C–OH, Ar), 157.41 (C–OH, Ar),
150.60 (Ar–C), 140.06 (Ar–C), 129.12 (Ar–Cmeta), 117.40
t
t
(Ar–C), 38.43 (CH2–NH), 34.07 (C–, Bu), 31.14 (CH3, Bu).
MS ESI(+): m/z 525 ([MH]+, 100%). UV-vis (DMSO): λmax/nm
(ε/M−1 cm−1) 317 (9791). M.p. 203 °C.
[NBu4]2 [LH4]. To a solution of LH6 (0.102 g, 0.194 mmol)
in dry methanol (10 mL), was added 2 eq of tetrabutylammo-
nium hydroxide (1
M solution in methanol, 389 μL,
1H NMR (300 MHz, CDCl3, 25 °C): δ/ppm 1.36 (s, 9H, Bu),
t
0.389 mmol) under N2 atmosphere. The colourless solution was
stirred under N2 atmosphere for one hour. Then, the solvent was
removed under vacuum yielding a white solid that was recrystal-
lised from the diffusion of dry Et2O into a saturated CH2Cl2
solution of [NBu4]2·[LH4]. The crystals were filtered and washed
several times with Et2O, then dried under vacuum to give pure
[NBu4]2·[LH4] in 80% yield (158 mg). Single crystals suitable
for X-ray crystallography were obtained in this manner. Elemen-
tal analysis: Calc. For C60H106N6O6: C 71.53, H 10.60, N 8.34;
Found: C 71.99, H 10.34, N 8.20. 1H NMR (300 MHz,
CD3CN): δ/ppm 12.82 (br s, 4H, NH), 7.97 (s, 4H, ArH),
2.91(s, 8H, CH2), 4.02 (s, 3H, CH3), 8.2 (s, 2H, ArH). 13C
t
NMR (75 MHz, CDCl3, 25 °C): δ/ppm 25.7, 31 (CH3, Bu),
t
34.77 (C–, Bu), 64.43 (CH3–O), 121.22 (Ar–C), 134.79 (Ar–
C
meta), 147.43 (Ar–C), 159.81(C–OMe, Ar), 160.4 (C(O)O),
169.98 (C(O)N). MS ESI(+): m/z 447 ([3H+]+, 100%).
N1,N3-Bis(2-aminoethyl)-5-(tert-butyl)-2-methoxyisophthalamide
(4). To a solution of ethylene diamine (0.4 g, 6.72 mmol,
0.448 mL) in 20 mL of CH2Cl2, was added dropwise a solution
of P (1.0 g, 2.24 mmol) in 90 mL of CH2Cl2 under N2 atmos-
phere at room temperature. The resulting suspension was stirred
12458 | Dalton Trans., 2012, 41, 12457–12467
This journal is © The Royal Society of Chemistry 2012

View Article Online
3.46–3.44 (m, 8H, CH2), 3.09–3.03 (m, 16H, CH2–[NBu4]),
1.63–1.52 (m, 16H, CH2–[NBu4]), 1.39–1.3 (m, 16H, CH2–
[NBu4]), 1.27 (s, 18H, tBu), 0.95 (t, 24H, CH3–[NBu4]).
1H NMR (300 MHz, DMSO-d6) δ/ppm 12.72 (br s, 4H, NH),
7.85 (s, 4H, ArH), 3.36–3.32 (br m, 8H, CH2), 3.18–3.13 (m,
16H, CH2–[NBu4]), 1.56–1.53 (m, 16H, CH2–[NBu4]),
(−): m/z 288 ([1]2−, 100%). UV-vis (CH2Cl2): λmax/nm (ε/M−1
cm−1): 310 sh (5544) 363 (6954), 464 sh (198), 545 (98).
[NBu4]2[2]. Elemental analysis: Calc. for C60H104N6O6-
Cu·C6H14·CH2Cl2: C 64.89, H 9.75, N 6.78; Found: C 64.45,
H 9.72, N 7.10. MS ESI(−): m/z 291 ([2]2−, 100%). UV-vis
(CH2Cl2): λmax/nm (ε/M−1 cm−1): 338 (12 536), 515 (186).
t
1.34–1.27 (m, 16H, CH2–[NBu4]), 1.22 (s, 18H, Bu), 0.93
(t, 24H, CH3–[NBu4]). 13C NMR (75 MHz, DMSO-d6, 25 °C):
δ/ppm 170 (C(O)NH), 167.97 (C(O)NH), 150.46 (Ar–Cpara),
129.05 (Ar–Cmeta), 128.8 (Ar–Cortho), 120.03 (Ar–CO), 57.5
Li2[1]. To a suspension of LH6 (0.1 g, 0.19 mmol) in dry
THF (7 mL), n-BuLi (2.5 M in hexane, 304 μL, 0.76 mmol) was
added dropwise under N2 at −78 °C. Then a solution of Ni-
(acac)2 (49 mg, 0.19 mmol) in 10 mL of dry THF, was added to
the reaction mixture, causing the colour of the solution to turn
intense orange. The reaction mixture was stirred overnight under
N2, filtered and evaporated under vacuum, yielding an orange
solid of Li2[1] in 80% yield (90 mg). 1H NMR (300 MHz,
CD3OD, 298 K): δ 11.27 (bs, 2H, NH), 8.12 (d, 3 Hz, 2H,
ArH), 7.92 (d, 3 Hz, 2H, ArH), 3.56 (s, 4H, CH2), 3.10 (bs, 4H,
CH2), 1.29 (s, 18H, tBu). 13C NMR (300 MHz, CD3OD,
298 K): δ/ppm: 170.58 (C(O)NH), 168.11 (C(O)NH), 164.68
(C–O, Ar), 136.73 (Ar–Cpara), 132.14 (Ar–Cmeta), 129.52 (Ar–
t
t
(CH2–[NBu4]), 38.49 (CH2), 33.24 (C–, Bu), 31.76 (CH3, Bu),
23.03 (CH2–[NBu4]), 19.18 (CH2–[NBu4]), 13.46 (CH3–
[NBu4]). 13C NMR (75 MHz, CD3CN, 25 °C): δ/ppm 169.86
(C(O)NH), 130.58 (Ar–Cmeta), 121.39 (Ar–C), 59.33 (CH2–
t
t
[NBu4]), 39.63 (CH3, Bu), 34.29 (C–, Bu), 32.23 (CH2), 24.32
(CH2, [NBu4]), 20.32 (CH2, [NBu4]), 13.8 (CH3, [NBu4]). MS
ESI(−): m/z 523 ([LH5]−, 100%), UV-vis (DMSO): λmax/nm
(ε/M−1 cm−1) 363 (14 463).
Complex synthesis
C
meta), 125.02 (Ar–Cortho), 120.84 (Ar–Cortho), 39.66 (CH2),
t
t
36.95 (C–, Bu), 32.09(CH3, Bu). MS ESI(−): m/z 289 ([1]2−
,
To a suspension of LH6 (0.1 g, 0.19 mmol) in dry methanol
(6 mL), 0.38 mL of a solution of [NBu4](OH) (1 M in CH3OH;
0.38 mmol) was added dropwise under N2 at room temperature.
The reaction mixture forms a clear colourless solution within
10 min. Then a solution of [Ni(OAc)2]·(H2O)4 (47 mg,
0.19 mmol) or [Cu(OAc)2]·(H2O) (38 mg, 0.19 mmol) in 10 mL
of dry MeOH, was added to the reaction mixture, causing the
colour of the solution to turn orange-red or light pink, respect-
ively. The reaction mixture was stirred for 1 h, after which time
0.38 mL (0.38 mmol) of a 1 M solution of [NBu4]·(OH) in
CH3OH was added. The solution was stirred for 1 h, filtered and
evaporated under vacuum yielding orange-red or pinkish oily
material, respectively. Crystallisation from cold acetonitrile pro-
duced orange crystals of [NBu4]2[1] in 87% yield (175 mg) and
crystallization from liquid–liquid diffusion of hexane in a
CH2Cl2 solution produced red crystals of [NBu4]2[2] in 84%
yield (170 mg). Both single crystals were suitable for X-ray
crystallography.
100%).
Na2[1]. To a suspension of LH6 (0.1 g, 0.19 mmol) in dry
DMF (6 mL), NaH (60% in oil, 0.03 g, 0.76 mmol) was added
under N2 at −78 °C. Then a solution of Ni(acac)2 (48.8 mg,
0.19 mmol) in 10 mL of dry DMF, was added to the reaction
mixture, causing the colour of the solution to turn intense
orange. The reaction mixture was stirred overnight under N2,
neutralised by addition of ethanol, filtered and evaporated under
vacuum yielding an orange solid of Na2[1] in 84% yield
1
(100 mg). H NMR (300 MHz, CD3OD, 298 K): δ 11.27 (bs,
2H, NH), 8.12 (d, 3 Hz, 2H, Ar H), 7.92 (d, 3 Hz, 2H, Ar H),
t
3.56 (s, 4H, CH2), 3.10 (bs, 4H, CH2), 1.29 (s, 18H, Bu).
13C NMR (300 MHz, CD3OD, 298 K): δ/ppm: 170.58 (C(O)
NH), 168.11 (C(O)NH), 164.68 (C–O, Ar), 136.73 (Ar–Cpara),
132.14 (Ar–Cmeta), 129.52 (Ar–Cmeta), 125.02 (Ar–Cortho),
t
120.84 (Ar–Cortho), 39.66 (CH2), 36.95 (C–, Bu), 32.09 (CH3,
tBu). MS ESI(−): m/z 289 ([1]2−, 100%).
[NBu4]2[1]. Elemental analysis: Calc. for C60H104N6O6Ni.
Physical methods
CH3OH: C 66.59, H 9.87, N 7.77; Found: C 64.40, H 9.47,
1
N 7.32. H NMR (300 MHz, CD3CN, 298 K): δ/ppm 11.09 (bs,
Elemental analyses of the compounds isolated in these studies
were accomplished using the chemistry departmental service
(BIU). ESI mass spectra were recorded on a Q-Tof micro (UK)-
2H, NH), 8.01 (d, 4J = 3 Hz, 2H, Ar–H), 7.81 (d, 4J = 3 Hz, 2H,
Ar–H), 3.43 (m, 4H, CH2), 3.10 (m, 16H, CH2–[NBu4]), 2.90
t
1
(s, 4H, CH2), 1.60 (m, 16H, CH2–[NBu4]), 1.32 (s, 18H, Bu),
micromass-Waters spectrometer. H and 13C NMR spectra were
1.31 (m, 16H, CH2–[NBu4]), 0.94 (t, 7.35 Hz, 24H CH3–
recorded on Bruker DPX300 NMR and Bruker Avance III 400
spectrometer operating at 300 and 400 MHz, respectively for H
1
1
[NBu4]). H NMR (300 MHz, CD3OD, 298 K): δ/ppm 11.30
(bs, 2H, NH), 8.13 (d, 4J = 3 Hz, 2H, Ar–H), 7.93 (d, 4J = 3 Hz,
2H, Ar–H), 3.54 (bs, 4H, CH2), 3.24 (m, 16H, CH2–[NBu4]),
3.11 (s, 4H, CH2), 1.63 (m, 16H, CH2–[NBu4]), 1.38 (m, 16H,
and 75 and 100.62 MHz, respectively for 13C. Chemical shifts
are reported in ppm downfield from tetramethylsilane, and coup-
ling constants (J) are reported in Hz. Resonance and structural
assignments were based on the analysis of coupling patterns,
including the 13C–1H coupling profiles obtained in bidimen-
sional HMBC and HSQC experiments, performed with standard
pulse programs. EPR spectra were recorded on a Bruker X-band
EMX spectrometer; all spectra were calibrated using DPPH
radical as an internal standard, and g-values were calculated
accordingly. Frozen solution EPR spectra were recorded at 77 K
t
CH2–[NBu4]), 1.29 (s, 18H, Bu), 1.00 (t, 7.35 Hz, 24H, CH3–
[NBu4]). 13C NMR (75 MHz, CD3CN, 298 K): δ/ppm 168.51
(C(O)NH), 166.75 (C(O)NH), 164.55 (C–O, Ar), 135.04 (Ar–
C
C
para), 131.10 (Ar–Cmeta), 128.12 (Ar–Cmeta), 126.09 (Ar–
ortho), 121.48 (Ar–Cortho), 59.29 (CH2–[NBu4]), 49.99 (CH2),
t
t
39.30 (CH2), 34.39 (C–, Bu), 32.11 (CH3, Bu), 24.33 (CH2,
[NBu4]), 20.33 (CH2, [NBu4]), 13.81 (CH3, [NBu4]). MS ESI
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12457–12467 | 12459

View Article Online
Table 1 Crystallographic data for [LH4(OMe)2], [NBu4]2[LH4], [NBu4]2[1], and [NBu4]2[2]
[LH4(OMe)2]·2H2O
30H40N4O6·2H2O
588.69
Monoclinic
P21/n
[NBu4]2·[LH4]
[NBu4]2[1]·0.5H2O
[NBu4]2[2]·0.5H2O
Empirical formula
M
Crystal system
Space group (no)
C
C60H106N6O6
1007.51
Monoclinic
P21/c
C28H30N4NiO6·2(C16H36N)·H2O
C28H30CuN4O6·2(C4H9)4N·0.5H2O
1080.20
Triclinic
ˉ
P1
1076.02
Triclinic
ˉ
P1
a/Å
b/Å
c/Å
α/°
12.4480(4)
6.6970(2)
19.3150(7)
90.00
106.2690(10)
90.00
1545.70(9)
2
1.265
10.1559(3)
17.4972(6)
16.9057(5)
90.00
100.921(10)
90.00
2949.73(16)
2
1.134
9.6113(9)
10.1921(9)
16.9872(15)
98.776(2)
100.154(2)
107.980(2)
1519.4(2)
1
9.3235(2)
10.3258(2)
16.9141(4)
98.869(10)
100.635(10)
108.337(10)
1479.29(6)
1
β/°
γ/°
V/Å3
Z
Dc/g cm−3
1.181
0.373
1.208
0.943
μ/mm−1
0.092
0.072
Reflections collected
Independent reflections (Rint
R1 (obs)
22 445
3726(0.0915)
0.0480
33 608
7107(0.1023)
0.0641
15 020
6945(0.0345)
0.0864
5724
3145(0.0247)
0.0596
)
wR2 (all)
0.1052
0.1546
0.280
0.1591
using a quartz finger dewar (Wilmad WG-816-Q) filled with
liquid nitrogen. All EPR spectra simulations were performed
using Bruker Simphonia software package. UV-vis spectra were
recorded on a Varian Cary 5000 UV-vis-NIR spectrophotometer.
The measurements were carried out using a quartz cuvette with
optical path length 0.1 cm. IR spectra were recorded on a
Nicolet iS10 FT-IR spectrometer using ATR accessory on ZnSe
crystal for powder samples pressed.
with fixed SOF 0.25. The crystallographic and refinement para-
meters for the studied compounds are given in Table 1. CCDC
numbers 883918–883921 contain the supplementary crystallo-
graphic data for this paper.†
DFT calculations
All theoretical calculations were performed with the ORCA
program package.14a Full geometry optimizations were carried
out for all complexes using the GGA functional BP8614b–d in
combination with the TZV/P14e basis set for all atoms and by
taking advantage of the resolution of the identity (RI) approxi-
mation in the Split-RI-J variant14f with the appropriate Coulomb
fitting sets.14g To properly model the interaction via H-bonding
in the complexes, the TZV/P++ basis set (the same basis set aug-
mented with diffuse functions on all atoms) was employed for
the relevant H, N and O atoms. Increased integration grids
(Grid4 in ORCA convention) and tight SCF convergence criteria
were used. Solvent effects were accounted for according to the
experimental conditions. For that purpose, we used the CH2Cl2
(ε = 9.08) solvent within the framework of the conductor-like
screening (COSMO) dielectric continuum approach.14h
g-Tensors and hyperfine coupling constants were obtained from
relativistic single-point calculations using the B3LYP functio-
nal.14i,j Scalar relativistic effects were included with ZORA
paired with the SARC def2-TZVP(-f) basis sets14k,l and the
decontracted def2-TZVP/J Coulomb fitting basis sets for all
atoms. Increased integration grids (Grid4 and GridX4 in ORCA
convention) and tight SCF convergence criteria were used in the
calculation. The integration grids were increased to an inte-
gration accuracy of 11 (ORCA convention) for the metal center.
Picture change effects were applied for the calculation of the
hyperfine tensors. Optical properties were also obtained from
single-point calculations using the hybrid functional B3LYP14i,j
and the TZV/P14e basis set. Electronic transition energies and
dipole moments for all models were calculated using time-
dependent DFT (TD-DFT)14m–o within the Tamm–Dancoff
approximation.14p,q To increase computational efficiency, the RI
X-ray crystallography
Single crystal X-ray diffraction data for compounds [LH4(OMe)2],
[NBu4]2[LH4] and [NBu4]2[1] were collected at 120 K on a
Bruker SMART 6000 CCD area detector diffractometer
equipped with a Cryostream (Oxford Cryosystems) open-flow
nitrogen cryostat using graphite-monochromated Mo-Kα radia-
tion (λ = 0.71073 Å). Data for compound [NBu4]2[2] were col-
lected at 100 K on a Bruker Proteum Pt-135 CCD area detector
diffractometer equipped with a Cobra (Oxford Cryosystems)
open-flow nitrogen cryostat using mirror-focused Cu-Kα radia-
tion (λ = 1.54188 Å) from a Bruker Microstar rotating anode
generator. All structures were solved by direct methods and
refined by full-matrix least-squares on F2 for all data using
SHELXTL12 and OLEX213 software. All non-disordered non-H
atoms were refined with anisotropic atomic displacement para-
meters. H atoms in structures [LH4(OMe)2] and [NBu4]2[LH4]
were located on a difference Fourier map and refined isotropi-
cally. H atoms in structures [NBu4]2[1] and [NBu4]2[2] were
placed in calculated positions and refined in a “riding mode”.
Disordered water oxygen atoms in the structure [LH4(OMe)2]
were refined isotropically with fixed SOF 0.6 and 0.4, and the
corresponding hydrogen atoms were restricted to have identical
thermal and positional parameters. Disordered n-Bu and tert-Bu
groups in the structure [NBu4]2[1] were refined with fixed SOF
0.7 and 0.3 and N–C and C–C bond lengths constrained to be
equal in both components; Ni atom and solvent water molecule
with fixed SOF 0.5. A terminal Me group in one of the NBu4
cations in [NBu4]2[2] is also disordered (fixed SOF 0.75 and
0.25), the solvent water molecule in this structure was refined
12460 | Dalton Trans., 2012, 41, 12457–12467
This journal is © The Royal Society of Chemistry 2012

View Article Online
approximation14r was used in calculating the Coulomb term and
at least 30 excited states were calculated in each case.
of the macrocycle in these two compounds is quite different. In
the structure of [NBu4]2[LH4], the dianionic macrocycle, each of
the two phenolate O atoms are involved in relatively strong15
O⋯HN intramolecular hydrogen bonds with the two adjacent
amide N–H (N⋯O 2.563 and 2.567 Å, both N–H⋯O angles are
equal to 151(3)°) creating four additional R11(6) H-bonded cycles
inside the main macrocycle (Fig. 2). These hydrogen bonds
maintain a planar orientation of the amide groups relative to the
aromatic rings and thus restrict the conformational flexibility of
the dianionic macrocycle. In contrast to [NBu4]2[LH4], only one
of the independent amide groups of the molecule [LH4(OMe)2]
forms a similar intramolecular H-bond with the oxygen atom of
the methoxy group (N⋯O 2.699(2) Å, N–H⋯O angle 136(2)°)
while the other amide group is H-bonded to a solvent water
molecule. It results in a significant deviation of the latter amide
group out of the plane of the aromatic ring; the corresponding
torsion angles around C2–C7 and C6–C9 bonds are equal to
127.0 and 12.3° respectively (Fig. 2). These different orien-
tations of the amide groups make the general shape of the macro-
cycle in [LH4(OMe)2] significantly less planar and consequently
more open than in [NBu4]2[LH4]. Indeed the distance between
the two aromatic planes in [LH4(OMe)2] of 3.90 Å is much
larger than the corresponding distance (1.88 Å) in the dianion of
[NBu4]2[LH4]. In fact the methoxy groups in molecule
[LH4(OMe)2] overlap each other with the O⋯O distance equal
to 3.904 Å (2.88 Å in [NBu4]2[LH4]). These conformational
differences of two closely related compounds emphasize the
potential flexibility of the macrocycle and the important role of
intramolecular hydrogen bonds.
Results and discussion
The macrocyclic compounds
The synthesis of new macrocyclic pro-ligand LH6 was success-
fully achieved in seven steps starting from 4-tert-butylphenol (1)
(Fig. 1). The latter was converted to the diacid 5-(tert-butyl)-
2-methoxyisophthalic acid (2) in three steps as reported.11 The
two carboxylic groups of 2 were then activated by N-hydroxy-
succinimide groups to provide the precursor 3. The latter was
reacted with 3 eq of ethylenediamine in CH2Cl2 to provide the
diamide compound 4 possessing two pendant amino arms. Com-
pound 4 was found to degrade with time (a few days) into a
brown oily material, and was therefore used as freshly made.
The diamide 4 was then condensed with 3 in diluted conditions
(to favour intramolecular ring closing) in presence of Et3N, at
room temperature under N2 for 72 h. The main product was
separated by column chromatography (eluent: ethyl acetate–
methanol 90 : 10) and identified as the macrocycle LH4(OMe)2.
Finally the straightforward demethylation of the two phenol
groups in LH4(OMe)2 using HBr in acetic acid yields the bis-
2-hydroxyisophtalamide macrocycle LH6 in 94% yield. The
corresponding di-phenolate salt [NBu4]2[LH4] was easily
obtained using two equivalents of [NBu4](OH), and has proved
to be stable in solution under N2 as well as in the solid state in
open air.
The three macrocyclic compounds, [LH4(OMe)2], LH6 and
1
[NBu4]2[LH4] have been fully characterised in solution by H,
13C NMR, MS and in the solid state by X-ray crystallography
for [LH4(OMe)2] and [NBu4]2[LH4].
NMR studies
[LH4(OMe)2], LH6 and [NBu4]2[LH4] have been characterised
by H and 13C NMR in DMSO-d6. The spectrum of LH6 dis-
plays a phenolic O–H resonance at 14.16 ppm that is character-
istic of O–H⋯OvC intramolecular H-bonding as observed in
salicylamide and 2-hydroxyisophtalamide compounds.16,17 This
1
X-ray crystallography
The molecular structures of [LH4(OMe)2] and [NBu4]2[LH4]
have been determined by X-ray crystallography (Fig. 2, Table 1).
Both molecules in crystals are located in special positions in
centres of symmetry and therefore in both molecules aromatic
fragments are parallel to each other. However, the conformation
Fig. 2 ORTEP representations of the molecular structures of
LH4(OMe)2 (left top/bottom) and [LH4]2− in [NBu4]2[LH4] (right top/
bottom) (atoms are shown with 30% thermal ellipsoids). Hydrogen
bonding interactions are represented with dotted lines. Only the N–H
hydrogen atoms are shown and the tetrabutylammonium cations have
been omitted for clarity.
Fig. 1 Synthetic route to LH6. (i) (see ref. 11); (ii) DCC, NHS,
dioxane, N2, 12 h, RT; (iii) 3 eq ethylenediamine, CH2Cl2, N2, 12 h, RT;
(iv) 3, Et3N, CH2Cl2, N2, 72 h, RT; (v) 33% HBr AcOH, reflux, 1 h;
(vi) [OH][NBu4] (1 M in MeOH), RT, 1 h, MeOH, N2.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12457–12467 | 12461

View Article Online
conformation is further confirmed by an amide N–H resonance
at 8.11 ppm, as expected for a non-hydrogen bonded amide. The
spectrum [LH4(OMe)2] also exhibits an amide N–H resonance in
the same range (8.77 ppm) indicating that the intramolecular
MeO⋯H–N H bond in crystal of [LH4(OMe)2] is not retained
in a DMSO-d6 solution which further emphasizes the flexibility
of the macrocycle. In contrast, the 1H NMR spectrum of the
di-phenolate salt [NBu4]2[LH4] displays an amide N–H reso-
nance at a much lower field (12.72 ppm) as expected for an
amide N–H intramolecularly hydrogen bonded to a phenolate
O atom.17 Thus, the (N–H⋯O−⋯H–N)2 biplanar conformation
revealed by the X-ray structure of [LH4]2− is preserved in solu-
tion. Interestingly, when the H NMR of [LH4]2− was recorded
in CD3OD, the N–H peak disappeared indicating a fast H/D
exchange.
1
Scheme 2
For all three macrocycles [LH4(OMe)2], LH6 and
[NBu4]2[LH4], a single peak is observed for each group of
protons, i.e., the meta-H (4H; 7.8–8 ppm), the ethylene CH2
The combined spectroscopic characterisation (EPR, NMR and
UV-vis) and DFT calculations strongly suggest that 12− and 22−
are isostructural mononuclear unsymmetrical intramolecularly
H-bonded complexes in the solid state as well as in solution
(Scheme 2).
t
(8H; 3–3.5 ppm) and the Bu (18H; 1.2–1.3 ppm) groups. This
indicates that, in all cases, that the macrocycle is symmetrical
and the two phenol-diamide units have the same environment.
Alternatively, rapid free rotation within the macrocycle would
also result in a single set of peak in the NMR spectrum.
However, if this is the case, this fluxional behaviour must be
rapid as the VT-NMR experiments performed for [NBu4]2[LH4]
in CD3CN showed no significant change within the temperature
range of 298–233 K (see ESI, Fig. SI1†). Analogous VT exper-
iments for [LH4(OMe)2] and LH6 have been hampered by the
low solubility of these compounds at low temperature.
X-ray crystallography
The molecular structures of [NBu4]2[1] and [NBu4]2[2] have
been determined by X-ray crystallography (Table 1, Fig. 3). The
X-ray structures confirm the di-anionic nature of the complexes
by the presence of two [NBu4]+ counter cations per complex.
Both compounds are isostructural (see Table 1) and contain one
metal ion per ligand. The macrocyclic ligand in these complexes
is planar in contrast to that in the structures of [LH4(OMe)2] and
[NBu4]2[LH4]. The two N2O2 compartments are virtually identi-
cal and both coordination sites are half-occupied by the metal
ions. This 0.5/0.5 occupancy is explained by the fact that the
two halves of the anion are symmetrically related over a centre
The [NBu4]2[1] and [NBu4]2[2] complexes
Many attempts have been made in order to synthesise the desired
dianionic dinuclear complexes [M2L]2− (M = Cu(II) or Ni(II)), in
which the two metal ions would be lodged in each (N2O2)4−
macrocyclic compartment. For that, strong bases such as
[NBu4](OH), NaH, or BuLi were reacted with LH6 in a 1 : 6
LH6/base ratio, prior the addition of the metal salt (1 : 2 LH6/
metal ion ratio). The reactions were either carried out in dry
MeOH when [NBu4](OH) and MII(acetate)2·(H2O)n were used,
or under dry and inert conditions using anhydrous THF when
NaH, BuLi, and anhydrous MII(acetate)2 salts were used. Sur-
prisingly, regardless of the reactions conditions used (room
temperature, reflux, in microwave oven (T = 70 °C), or hydro-
thermal conditions (pressure tube, oven 90 °C)), only the corres-
ponding mononuclear complexes [X]2[MLH2] (M = Cu2+,
Ni2+; X = Li+, Na+, [NBu4]+) were obtained. When [NBu4](OH)
was used, the supplementary excess of metal ion is recovered as
M(OH)2 which precipitated out. Similar results were obtained
when [NBu4]2[LH4] was used instead of LH6. The nickel com-
plexes 12− could be obtained quantitatively with Li+, Na+
counter cations; and these were characterised by NMR and MS
showing that these are analogous to [NBu4]2[1] (see experi-
mental part). Finally, the [NBu4]2[MLH2] salts were synthesised
in a clean and quantitative manner by using only 4 eq of base
and an equimolar amount of metal salts. Thus, LH6 reacts with
MII(acetate)2·(H2O)6 (M = Ni, Cu) in a 1 : 1 ratio in the presence
of 4 eq of [NBu4](OH) in methanol to afford the salts
[NBu4]2[1] (M = Ni) and [NBu4]2[2] (M = Cu) respectively.
ˉ
of symmetry in the P1 (Z = 1) situation.
The difference between occupied and unoccupied compart-
ments is reflected by relatively large thermal displacement para-
meters of the coordinating atoms of the macrocycle. Presumably,
the N–H⋯O intramolecular H-bonds, similar to those found
in the structures of [LH4(OMe)2] and [NBu4]2[LH4], are
present in the unoccupied compartment and the corresponding
N⋯O distances (av. 2.630 in [NBu4]2[1] and 2.667 Å in
[NBu4]2[2]) are similar in all studied compounds, where such
H-bonds exist. Thus, even if asymmetry of the compound could
Fig. 3 ORTEP representation of 2− in [NBu4]2[2] (the displacement
ellipsoids are shown at the 30% probability level).
12462 | Dalton Trans., 2012, 41, 12457–12467
This journal is © The Royal Society of Chemistry 2012

View Article Online
not be evidenced by the X-ray analysis, the structures confirm
the tetra-anionic N2O2 coordination of the metal ion through O
phenolate and N amidate donor atoms in a square planar geome-
try (in both complexes the deviation of metal ion from the plane
formed by the four coordinating atoms is less than 0.05 Å). The
average M–O and M–N bond distances are equal to 1.795(4) and
1.879(5) Å in 12− and 1.802(4) and 1.942(4) Å in 22−
(respectively).
for the hyperfine coupling of the unpaired electron with the two
coordinated 14N nuclei spins (I = 1). The spectrum was success-
fully simulated with an isotropic constant (A14N) of 13 G, confi-
rming the N2 amidate coordination to the CuII ion in solution
(Fig. 4). The X-band EPR powder spectra of [NBu4]2[2] at room
temperature and at 77 K exhibit a broad unresolved peak at ca.
g = 2.1. The lack of resolution is characteristic of the involve-
ment of intermolecular interaction between the paramagnetic
centres. However, when a magnetically diluted powder of
[NBu4]2[2] (2%) with the isostructural diamagnetic [NBu4]2[1]
compound (98%) was examined, the RT powder EPR spectrum
showed a well-resolved anisotropic Cu(II) axial signal, possessing
a (dx2−y2)1 ground state (S = 1/2) with gzz ≫ gxx ∼ gyy > ge.18 As
shown in Fig. 4, the latter spectrum is similar to that of
[NBu4]2[2] in frozen CH2Cl2 solution at 77 K (best simulated
parameters: gx = 2.057, gy = 2.054, gz = 2.197 with ACu : Axx
30 G, Ayy 30 G and Azz 210 G and AisoN 15 G; see Fig. SI2†).
This observation clearly indicates that the planar structure of
[NBu4]2[2] in the solid state as revealed by its X-ray structure, is
retained in a CH2Cl2 solution. This is further confirmed by the
relatively low gzz of 2.20 and the relatively high Azz{63,65Cu} =
210 G simulated values which are typically observed for Cu(II)–
N2O2 complexes with a square planar geometry.19
In solution state, the presence of only one metal ion in 12−
and 22− is unambiguously confirmed in solution by (a) the mass
spectroscopy ES(−) displaying a peak at m/z = 288 and 291,
respectively corresponding to {MLH4}2−, (b) the observation of
a typical Cu(II) EPR signal in 22− and (c) the observation of a
well-resolved unsymmetrical 1H NMR spectrum for 12−
.
EPR studies
As expected for a single CuII ion (d9, S = 1/2), the complex
[NBu4]2[2] is EPR-active in the solid state and in solution at RT
or at 77 K. The room-temperature X-band EPR spectrum of
[NBu4]2[2] in a dichloromethane solution displays a typical iso-
tropic four-line spectrum centred at giso = 2.105, with an isotro-
pic superhyperfine coupling constant A63,65Cu of 110 G with the
63,65Cu nuclei spin (I = 3/2); see Fig. 4. In addition, further
hyperfine splitting to a five-line pattern was observed, accounting
NMR studies
In contrast to [NBu4]2[2], [NBu4]2[1] is diamagnetic (S = 0) as
expected for a d8 square planar Ni(II) ion, and exhibits a well
1
resolved H NMR spectrum in CD3CN and DMSO-d6 (Fig. 5).
The 1H NMR spectrum of [NBu4]2[1] is strikingly different from
that of the macrocyclic compounds [LH4(OMe)2], LH6 and
[NBu4]2[LH4], in that the spectrum displays a dissymmetry in
the dianionic 12− core. Indeed, both the two aromatic meta-
protons, as well as the ethylene protons, exhibit two sets of reso-
nances revealing a dissymmetry, i.e., the two compartments of
the macrocycle are no longer equivalent. Specifically, while one
set of four ethylene protons appears as
a multiplet at
3.43–3.41 ppm (Fig. 5, CH2 b), in the same region as observed
for [LH4(OMe)2], LH6 and [NBu4]2[LH4]; the position of the
other four ethylene protons resonances appears significantly
Fig. 4 X-band EPR spectra of (top) [NBu4]2[2] (ca. 1 mM) in CH2Cl2
recorded at room temperature (blue line) and its simulated spectrum (red
line); and of (bottom) powder of [NBu4]2[2] (2%) in [NBu4]2[1] (98%)
at 77 K (black line) and frozen solution of [NBu4]2[2] (ca. 1 mM) in
CH2Cl2 recorded at 77 K (red line).
Fig. 5 Room-temperature 1H NMR (CD3CN, 300 MHz) spectrum of
[NBu4]2[1].
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12457–12467 | 12463

View Article Online
shifted at higher field, i.e. displaying a single peak at 2.9 ppm
(Fig. 5, CH2 a). The latter upfield shift is indicative of
N−-amidate ligation to the Ni(II) ion; as already observed in a
recently reported di-nickel(II) complex, [Ni2NHOHL2], bearing
NO2− salicylamidate ligands with a coordinated ethanolamine
arm (for which the four ethylene protons resonances occur at
2.32–2.54 ppm in CD3CN).20 Thus, in 12−, one side of the
macrocycle contains a Ni(II)–N2O2 dianionic core while the
tetra-anionic ligand, Matsumoto et al. observed a similar band
but weaker in bis-salicylamide Ni(II) and Cu(II) dianionic assem-
blies of Li+ or K+.6c,d The weakening of the visible band in 12−
and 22− can therefore not be fully explained by the presence of
the amidate ligation, and may be the result of H-bonding effects
particularly in pulling electron density from the coordinated
phenolate ring. The origin of these bands as well as the
H-bonding effect have been further correlated by TD-DFT calcu-
lations (vide infra).
1
other side is vacant. Another striking feature revealed by the H
NMR is the observation of an NH resonance (integrating for 2H)
at 11.10 ppm that clearly indicates that two N–H protons are
present and are involved in intramolecular NH⋯O H-bonds with
the two coordinated phenolates (Fig. 5). It is worth noting that
the N–H chemical shift is somewhat lower than that of the di-
phenolate salt [NBu4]2[LH4] (12.82 (bs, 4H, NH) (CD3CN),
12.72 (bs, 4H, NH) (DMSO)) and that of recently characterised
H-bonded mono-salicylamidate compounds [NBu4][NHOHL] and
[NBu4][NHMeL] in CD3CN: δ 13.90 and 13.20 ppm.17 This may
indicate that the phenolate O−⋯H–N H-bonding is weakened
when the phenolate is coordinated to the Ni(II).
Nevertheless, strikingly, the H-bonding interaction appears to
be relatively strong as it was noticed that the N–H resonance
remains visible for days at room temperature in CD3OD indicat-
ing that the proton did not exchange with deuterium. The same
observation was observed for sodium and lithium salts Na2[1]
and Li2[1], and contrasts drastically with the fast N(H/D)
exchange of [LH4]2− in CD3OD, despite the presence of similar
the O−⋯H–N H-bonding. This observation may indicate large
differences in flexibilities and N–H accessibilities between
[LH4]2− and 12−; the N–H being confined in a rigid square
planar geometry for the latter, restricting the rotation of the un-
coordinated N2O2-pocket, therefore preventing easy N(H/D)
exchange. The N(H/D) exchange in 12− in CD3OD could only
be observed at higher temperature, i.e., after 104 min at 323 K,
85% of the NH signal had disappeared (see Table SI1†).
DFT calculations
In order to validate our experimental data and gain more infor-
mation on the electronic structure of 12− and 22−, we have
performed DFT calculations. The optimised geometries of 12−
and 22− are depicted in Fig. 6 and the selected geometrical para-
meters are reported in Tables SI2–3.† In both geometrically opti-
mised structures, the metal ions (Ni(II) or Cu(II)) are coordinated
by two phenolate O–(O1, O2) and two amidate N–(N1, N2)
atoms, with Ni–O1: 1.871 and Ni–O2: 1.862 Å and Ni–N1:
1.862 and Ni–N2: 1.862 Å, respectively; Cu–O1: 1.928 Å and
Cu–O2: 1.926 and Cu–N1: 1.929 Å and Cu–N2: 1.929 Å,
respectively. These bond lengths are in the range of those
observed for Ni(II) and Cu(II) N2O2-bis-salicylamidate
complexes.6
The optimised geometry of 12− features an almost perfect
square planar geometry around the Ni(II) ion as observed in the
X-ray structure of 12−. Thus, the angle between the O1–Ni–N1
and O2–Ni–N2 planes is 2.8°, and the O1–Ni–O2, O1–Ni–N1,
Such H-bonding may be responsible for the great stability of
the complex and the difficulties encountered in inserting a
second metal ion in the vacant compartment.
Thus, in 12− the two phenolate groups are both coordinated to
the Ni(II) ion and H-bonded to the adjacent amide N–H, which
presumably maintained a planar structure as observed from the
X-ray data in the solid state and from the EPR in solution for the
analogous Cu(II) complex 22−
.
The compounds [NBu4]2[1] and [NBu4]2[2] are weakly
coloured in solution and in the solid state (light pink and light
orange respectively). Indeed, while in the UV region, 12− and 22−
display intense π–π* ligand-based transitions at 363 nm (ε =
6954 M−1 cm−1) and 338 nm (ε = 12 536 M−1 cm−1), respect-
ively; in the visible region, only very weak bands are observed at
464 nm (ε = 198 M−1 cm−1) and 545 nm (ε = 98 M−1 cm−1) for
12− and at 515 nm (ε = 186 M−1 cm−1) for 22− (see Fig. SI3†).
This is somewhat surprising and contrasts with square Cu(II) and
Ni(II) salen complexes and other M–N2O2 phenol–Schiff base
complexes which commonly exhibit a relatively intense brown-
purple colour due to a phenolate-to-M(II) charge-transfer typi-
cally around 500–600 nm (ε = 500–1000 M−1 cm−1). Interest-
ingly, whilst Thomas et al.6b observed a band at 515 nm (ε =
600 M−1 cm−1) for a discrete mononuclear square planar dianio-
nic [NBu4]2 [Ni(II)L] complex bearing a bis-salicylamide
Fig. 6 DFT optimised structures of [1]2− (top) and [2]2− (bottom) and
selected bond distances. Color scheme: Cu: light brown; Ni: purple; O:
red; N: dark blue; C: green and H: white.
12464 | Dalton Trans., 2012, 41, 12457–12467
This journal is © The Royal Society of Chemistry 2012

View Article Online
O2–Ni–N2 and N1–Ni–N2 angles differ only slightly from 90°.
Unlike the nickel complex, the optimized geometry of 22− fea-
tures a slight deviation from planarity with an angle between the
O1–Cu–N1 and O2–Cu–N2 planes of 13.4°; the O1–Cu–O2,
O1–Cu–N1, O2–Cu–N2 and N1–Cu–N2 angles differ only
slightly from 90°.
giso of 2.072. These values are typical for square planar Cu(II)
complexes, and are close to the experimental data obtained for
22−(gx = 2.057, gy = 2.054, gz = 2.197). Moreover, the calculated
63,65Cu hyperfine coupling constants of Amin = 32 G, Amid
=
35 G, Amax = 251 G and Aiso = 106 G; are in good agreement
with the experimental data obtained (Axx = 30 G, Ayy = 30 G and
Azz = 210 G). Finally the isotropic nitrogen hyperfine coupling
constants calculated to be 15 G for N1 and N2, match well with
Both geometrically optimised structures feature two NH⋯O
intramolecular H-bonds between the coordinated phenolate and
the adjacent N–H amide; the calculated O1⋯H1 and O2⋯H2
distances of 1.731 Å and 1.749 Å (for 12−) and 1.703 and
1.700 Å (for 22−) indicating a relatively strong H-bonding inter-
action. These interactions presumably stabilise the structure and
may explain the difficulties in generating and isolating the dinuc-
lear complex. This corroborates well with the energetically
demanding deuterium exchange observed experimentally by
the experimentally determined Aiso of 15 G obtained for 22−
.
These values are larger than those observed for Cu(II)–salen
complexes, and reflect strong ligation of the amidate-N atom as
supported by the non-negligible spin populations (0.13) found at
both N centres.
TD-DFT calculations were undertaken in order to rationalise
the nature of the transitions observed in the visible region for 12−
and 22−. The two bands experimentally observed for 12− at
464 nm and 545 nm are well reproduced by the calculations
revealing bands at 552 nm (βHOMO → βLUMO+2, f = 0.007)
and 452 nm (βHOMO−2 → βLUMO+2, f = 0.012), correspond-
ing to MLCT transitions. Both donor orbitals are essentially
metal-based while the acceptor one is mainly ligand-based (see
Fig. 8). For the complex 22−, only one transition could be calcu-
lated at 460 nm (βHOMO → βLUMO+2, f = 0.053) and matches
with the single band at 515 nm observed experimentally. The
donor orbital is also essentially metal-based while the acceptor
one is spread over the ligand unit LH2 (see Fig. 9).
NMR for the NH proton in 12−
.
The Highest Occupied Molecular Orbital (HOMO) of 12− is
depicted in Fig. 7 and is mainly a metal-centred orbital. The
HOMO of 12− displays 42% nickel character, 22% nitrogen char-
acter and 12% oxygen character. The substantial contribution of
the N atoms confirm the strong amidate ligation while the
O atom contribution to the HOMO is almost twice as low. This
points out that, in contrast to Ni–salen complexes which exhibit
a mainly ligand-based HOMO with significant participation of
the phenolate groups, the strong N-amidate donation seems to
strongly influence the nature of the HOMO. The strong contri-
bution of the Ni dxz orbital to the HOMO of 12− suggests that
the oxidation of the complex 12− should occur at the metal
centre, and possibly leading to oxidised species with a stronger
Ni(III) character than that of Ni–salen oxidised species that are
best described as Ni(II)–phenoxyl radicals.6b
Interestingly, the intensity of the calculated transitions for 12−
and 22− appear to be four times smaller than those calculated for
The spin density as well as the localised SOMO of 22− are
shown in Fig. 7. The spin density at the copper centre is found
to be 0.57, and the remaining spin density is equally distributed
between the neighbouring oxygen and nitrogen atoms (O (0.08)
and N (0.13)). Not surprisingly, the SOMO of 22− is a metal-
centred orbital featuring a 60% contribution of the Cu dx2−y2
orbital, as expected from ligand field theory for a square planar
Cu(II) complex, and as correlated by the experimental EPR data.
The DFT calculated EPR parameters for 22− indicate an axial
signal with g⊥ (gmin = 2.038, gmid = 2.041) and g = 2.138 and a
k
Fig. 8 TD-DFT assignment of the visible bands in 12−
.
Fig. 7 HOMO orbital of 12− (top left), SOMO orbital of 22− (bottom
left) and spin density plot of 22− (right).
Fig. 9 TD-DFT assignment of the visible bands in 22−
.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12457–12467 | 12465

View Article Online
analogous complexes containing only one N2O2 compartment,
i.e. without N–H⋯O H-bonding. This observation correlates the
hypothesis that H-bonding interaction decreases the intensity of
the visible transitions.
Thus, overall the geometrically optimised structures reproduce
well the experimental data, allowing an accurate description of
the geometric and electronic structures of the dianionic species
12− and 22− to be provided.
be used as complex ligand for further ligation of cationic entities
such as transitions metals and lanthanide ions. They can exist in
stable higher valence states permitting the studies of their elec-
tronic structures. We are currently pursuing our research work in
these directions.
Notes and references
1 (a) P. G. Cozzi, Chem. Soc. Rev., 2004, 33, 410; (b) A. D. Cort, P. De
Bernardin, G. Forte and F. Y. Mihan, Chem. Soc. Rev., 2010, 39, 3863.
2 (a) L. Baldini, A. Casnati, F. Sansone and R. Ungaro, Chem. Soc. Rev.,
2007, 36, 254; (b) D. T. Schühle, J. A. Peters and J. Schatz, Coord.
Chem. Rev., 2011, 255, 2727.
3 J. Xu, T. M. Corneillie, E. G. Moore, G.-L. Law, N. G. Butlin and
K. N. Raymond, J. Am. Chem. Soc., 2011, 133, 19900 and references
therein.
4 (a) H. Okawa, H. Furutachia and D. E. Fenton, Coord. Chem. Rev., 1998,
174, 51; (b) A. J. Atkins, D. Black, A. J. Blake, A. Marin-Becerra,
S. Parsons, L. Ruiz-Ramirez and M. Schröder, Chem. Commun., 1996,
457.
5 (a) D. L. Popescu, A. Chanda, M. Stadler, F. Tiago de Oliveira,
A. D. Ryabov, E. Muenck, E. L. Bominaar and T. J. Collins, Coord.
Chem. Rev., 2008, 252, 2050; (b) T. J. Collins, Acc. Chem. Res., 1994,
27, 279.
Concluding remarks
Herein, we have reported the synthesis of the unprecedented
2-hydroxyisophtalamide macrocycle LH6. The new synthetic
route described is versatile and facile, and may be easily
extended with various diamine bridges. The results of solid state
and solution studies on [LH4(OMe)2], LH6 and [NBu4]2[LH4]
have indicated that the macrocycle can adopt specific confor-
mations, which are influenced by H-bonding interactions as well
as the deviation of the amide carbonyl with respect to the phenol
plane. For instance, the di-phenolate macrocycle [LH4]2−, in
which both phenol-diamide units are maintained planar by a
N–H⋯O−⋯H–N H-bonding, possesses a much smaller cavity
than that of [LH4(OMe)2] in which two of the carbonyl amide
groups are significantly twisted out of the phenol plane, allowing
opening of the cavity. In LH6, the NMR data has indicated an
OH⋯O H-bonding indicating that the N–H groups are pointing
away from the macrocycle. Strikingly, the NMR spectra of these
macrocycles are all symmetrical exhibiting a single peak for all
CH2, NH, and meta-H. Though, this may be indicative of rapid
fluxional within the cavity ring; low temperature NMR experi-
ments in the range of 298–233 K did not show any significant
changes.
The new macrocycle LH6 was designed to accommodate two
metal centres with a N2O2 amidate-phenolato tetra-anionic
chelating unit. However, it turned out, that, in our hands, the
insertion of two metal ions (at least with Cu(II) and Ni(II) ions)
was problematic, and all attempts made have resulted in the for-
mation of the mononuclear dianionic [M(LH2)]2− complexes
designated as 12− (Ni(II)) and 22− (Cu(II)). The combined infor-
mation gathered by the X-ray, EPR, NMR, UV-vis and DFT cal-
culations are consistent with both complexes being square planar
MII–N2O2 dianionic complexes in which the phenolate O atoms
are being both coordinated to Ni(II) and H-bonded to the NH
amide of the vacant compartment. This intramolecular H-bond
has appeared to be relatively strong and rigid as indicated by an
N(H/D) exchange in CD3OD which occurs efficiently only at
elevated temperature (i.e. 323 K). Thus, the unique H-bonding
feature in 12− and 22− confer these complexes stability and rigid-
ity, and presumably plays a role in preventing the formation of
the putative dinuclear compound. Indeed, the H-bonding (a)
stabilises the negatively charged mononuclear complexes; (b)
maintains a planar and rigid geometry of the macrocycle render-
ing the amide N–H function hardly accessible; and (c) makes the
deprotonation of the amidic proton more difficult by increasing
its pKa. All of these factors contribute to rendering the pre-
paration of the dinuclear complex somewhat challenging.
6 (a) F. C. Anson, T. J. Collins, T. G. Richmond, B. D. Santarsiero,
J. E. Toth and B. G. R. T. Treco, J. Am. Chem. Soc., 1987, 109, 2974;
(b) O. Rotthaus, O. Jarjayes, F. Thomas, C. Philouze, C. P. Del Valle,
E. Saint-Aman and J. L. Pierre, Chem.–Eur. J., 2006, 12, 2293;
(c) Y. Sunatsuki, R. Hirata, Y. Motoda, M. Nakamura, N. Matsumoto and
F. Kai, Polyhedron, 1997, 16, 4105; (d) Y. Sunatsuki, M. Mimura,
H. Shimada, F. Kai and N. Matsumoto, Bull. Chem. Soc. Jpn., 1998, 71,
167.
7 C. A. Jimenez, J. B. Belmar, J. Alderete, F. S. Delgado, M. Lopez Rodri-
guez, O. Pena, M. Julve and C. Ruiz-Perez, Dalton Trans., 2007, 2135.
8 (a) Y. Sunatsuki, T. Matsuo, M. Nakamura, F. Kai, N. Matsumoto and
J. P. Tuchagues, Bull. Chem. Soc. Jpn., 1998, 71, 2611; (b) Y. Sunatsuki,
H. Shimada, T. Matsuo, M. Nakamura, F. Kai, N. Matsumoto and N. Re,
Inorg. Chem., 1998, 37, 5566; (c) Y. Sunatsuki, M. Nakamura,
N. Matsumoto and F. Kai, Bull. Chem. Soc. Jpn., 1997, 70, 1851;
(d) H. Z. Kou, Y. T. Wang, W. X. Luo, Q. W. Xie, J. Tao, A. L. Cui and
D. Z. Shen, Cryst. Growth Des., 2008, 8, 3908; (e) J. P. Costes, F. Dahan,
B. Donnadieu, M. J. Rodriguez Douton, M. I. Fernandez Garcia,
A. Bousseksou and J. P. Tuchagues, Inorg. Chem., 2004, 43, 2736;
(f) S. Osa, Y. Sunatsuki, Y. Yamamoto, M. Nakamura, T. Shimamoto,
N. Matsumoto and N. Re, Inorg. Chem., 2003, 42, 5507;
(g) M. Nakamura, M. Kawachi, C. Konatsu, S. Osa, T. Tarui,
N. Matsumoto and N. Re, Chem. Lett., 2003, 32, 936; (h) R. J. Tao,
F. A. Li, S. Q. Zang, Y. X. Cheng, Q. L. Wang, J. Y. Niu and D. Z. Liao,
Polyhedron, 2006, 25, 2153; (i) R. J. Tao, F. A. Li, Y. X. Cheng,
S. Q. Zang, Q. L. Wang, J. Y. Niu and D. Z. Liao, Inorg. Chim. Acta,
2006, 359, 2053.
9 (a) T. Kido, S. Nagasato, Y. Sunatsuki and N. Matsumoto, Chem.
Commun., 2000, 2113; (b) J. P. Costes and F. Dahan, C. R. Acad. Sci.,
Ser. IIc: Chim., 2001, 4, 97; (c) J. P. Costes, S. Shova, J. M. C. Juan and
N. Suet, Dalton Trans., 2005, 2830; (d) V. Gomez, L. Vendier,
M. Corbella and J. P. Costes, Eur. J. Inorg. Chem., 2011, 2653;
(e) S. Osa, T. Kido, N. Matsumoto, N. Re, A. Pochaba and J. Mrozinski,
J. Am. Chem. Soc., 2004, 126, 420; (f) T. Kido, Y. Ikuta, Y. Sunatsuki,
Y. Ogawa, N. Matsumoto and N. Re, Inorg. Chem., 2003, 42, 398;
(g) J. P. Costes, M. Auchel, F. Dahan, V. Peyrou, S. Shova and
W. Wernsdorfer, Inorg. Chem., 2006, 45, 1924; (h) T. Hamamatsu,
K. Yabe, M. Towatari, S. Osa, N. Matsumoto, N. Re, A. Pochaba,
J. Mrozinski, J. L. Gallani, A. Barla, P. Imperia, C. Paulsen and
J. P. Kappler, Inorg. Chem., 2007, 46, 4458; (i) J. P. Costes, L. Vendier
and W. Wernsdorfer, Dalton Trans., 2010, 39, 4886; ( j) J. P. Costes,
S. Shova and W. Wernsdorfer, Dalton Trans., 2008, 1843;
(k) T. Hamamatsu, N. Matsumoto, N. Re and J. Mrozinski, Chem. Lett.,
2009, 38, 762.
10 (a) H. Okawa, M. Honda and S. Kida, Chem. Lett., 1972, 1027;
(b) S. Brooker, D. J. de Geest and G. S. Dunbar, Inorg. Chim. Acta,
1998, 282, 222.
Nevertheless, these new mononuclear dianionic compounds
have interesting features that can be exploited further: they can
11 C. J. Fahrni and A. Pfaltz, Helv. Chim. Acta, 1998, 81, 491.
12466 | Dalton Trans., 2012, 41, 12457–12467
This journal is © The Royal Society of Chemistry 2012

View Article Online
12 G. M. Sheldrick, Acta Crystallogr., 2008, 64, 112.
13 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and
H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339.
109, 8218; (o) R. Bauernschmitt and R. Ahlrichs, Chem. Phys. Lett., 1996,
256, 454; (p) S. Hirata and M. Head-Gordon, Chem. Phys. Lett., 1999,
314, 291; (q) S. Hirata and M. Head-Gordon, Chem. Phys. Lett., 1999,
302, 375; (r) F. Neese and G. Olbrich, Chem. Phys. Lett., 2002, 362, 170.
15 G. A. Jeffrey, An Introduction to Hydrogen Bonding, Oxford University
Press, Oxford, 1997.
16 (a) D. Kanamori, T. A. Okamura, H. Yamamoto and N. Ueyama, Angew.
Chem., Int. Ed., 2005, 44, 969; (b) D. Kanamori, A. Furukawa, T. Okamura,
H. Yamamoto and N. Ueyama, Org. Biomol. Chem., 2005, 3, 1453.
17 R. Wanke, L. Benisvy, M. L. Kuznetsov, M. F. C. Guedes da Silva and
A. J. L. Pombeiro, Chem.–Eur. J., 2011, 17, 11882.
18 F. E. Mabbs and D. Collison, Electron Paramagnetic Resonance of
d-Transition Complexes, Elsevier, Amsterdam, 1992, p. 405.
19 G. M. Zats, H. Arora, R. Lavi, D. Yufit and L. Benisvy, Dalton Trans.,
2011, 40, 10889.
20 L. Benisvy, R. Wanke, M. F. C. Guedes da Silva and A. J. L. Pombeiro,
Eur. J. Inorg. Chem., 2011, 2791.
14 (a) F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2012, 2, 73;
(b) J. P. Perdew, Phys. Rev. B, 1986, 33, 8822; (c) J. P. Perdew, Phys. Rev.
B, 1986, 34, 7406; (d) A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys.,
1988, 38, 3098; (e) A. Schäfer, C. Huber and R. J. Ahlrichs, Chem.
Phys., 1994, 100, 5829; (f) F. Neese, J. Comput. Chem., 2003, 24, 1740;
(g) F. Weigend, Phys. Chem. Chem. Phys., 2006, 8, 1057; (h) A. Klamt
and G. J. Schürmann, J. Chem. Soc., Perkin Trans. 2, 1993, 799;
(i) A. D. Becke, J. Chem. Phys., 1993, 98, 1372; ( j) C. T. Lee, W. T. Yang
and R. G. Parr, Phys. Rev. B, 1988, 37, 785; (k) D. A. Pantazis,
X. Y. Chen, C. R. Landis and F. Neese, J. Chem. Theory Comput., 2008, 4,
908; (l) A. D. Pantazis and F. Neese, J. Chem. Theory Comput., 2009, 5,
2229; (m) M. E. Casida, in Recent Advances in Density Functional
Methods, ed. D. P. Chong, World Scientific, Singapore, 1995;
(n) R. E. Stratmann, G. E. Scuseria and M. J. Frisch, J. Chem. Phys., 1998,
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12457–12467 | 12467