Synthesis and characterization of new mono-, di-, and trinuclear copper(II) triethanolamine-carboxylate complexes

Article abstract of DOI:10.1002/zaac.200500204

A series of triethanolamine copper(II) carboxylate complexes have been synthesized and characterized both structurally and spectroscopically. The structures of the compounds and the nuclearity of the resulting coordination complexes are influenced both by the stoichiometry of the reagents and the steric and electronic influences of the anionic ligands present in the copper coordination spheres. The complexes can be prepared in high yields and purity as highly crystalline solids. Herein we report the spectroscopic and structural data for [NaCu(CF₃COO)₃(TEAH₃)] (1), [Cu(C ₆H₅COO)(TEAH₂)]₂·2H ₂O (2), [Cu₃(CH₃COO)₄(TEAH ₂)₂] (3), [Cu₃(HCOO)₄(TEAH ₂)₂] (4) and [Cu₃(CF₃OO) ₂(TEAH₂)₂-(CH₃OH) ₂][CF₃CO₂]₂ (5) (TEAH₃ = triethanolamine (N(CH₂CH₂OH)₃); [TEAH ₂]^- = mono-deprotonated ion; [TEAH]^2- = doubly deprotonated ion). Two approaches to these compounds were employed:direct reaction of copper(II) carboxylates with TEAH₃ and reaction of the tetrameric compound [{Cu(TEAH)}₄] (6) with copper(II) carboxylates. Compound 1 is polymeric in the solid state, being a double salt of sodium trifluoroacetate and (triethanolamine)copper(II) trifluoroacetate, while 2 exists as a simple dimer. Compounds 3, 4 and 5 are trinuclear complexes with Cu₃ chains connected by bridging carboxylate and triethanolamine ligands. 3 was characterized as three crystal modifications: an unsolvated form and as a methanol solvate 3.2MeOH that was found to exist in both monoclinic and orthorhombic forms. 4 was also found to exist as a methanol solvate, 4.1.70 MeOH. 5 shows coordination isomerism compared to compounds 3 and 4 in that two of the carboxylate ligands in the metal coordination spheres on the end copper atoms have been replaced by solvent methanol. The magnetic behavior of the di- and trinuclear species has been probed and reveals that at low temperatures the metal centers exhibit strong ferromagnetic coupling, which, on warming, switches to simple paramagnetic behavior.

Full text of DOI:10.1002/zaac.200500204

Synthesis and Characterization of New Mono-, Di-, and Trinuclear Copper(II)
Triethanolamine-Carboxylate Complexes
Raphael M. Escovar, John H. Thurston, Teyeb Ould-Ely, Arvind Kumar, and Kenton H. Whitmire*
Houston, TX / USA, Rice University, Department of Chemistry
th
Received April 28 , 2005.
th
Dedicated to Professor Herbert W. Roesky on the Occasion of his 70 Birthday
Abstract. A series of triethanolamine copper(II) carboxylate com-
plexes have been synthesized and characterized both structurally
and spectroscopically. The structures of the compounds and the
nuclearity of the resulting coordination complexes are influenced
both by the stoichiometry of the reagents and the steric and elec-
tronic influences of the anionic ligands present in the copper coor-
dination spheres. The complexes can be prepared in high yields and
purity as highly crystalline solids. Herein we report the spectro-
polymeric in the solid state, being a double salt of sodium trifluoro-
acetate and (triethanolamine)copper(II) trifluoroacetate, while 2
exists as a simple dimer. Compounds 3, 4 and 5 are trinuclear com-
3
plexes with Cu chains connected by bridging carboxylate and tri-
ethanolamine ligands. 3 was characterized as three crystal modifi-
cations: an unsolvated form and as a methanol solvate 3·2MeOH
that was found to exist in both monoclinic and orthorhombic
forms. 4 was also found to exist as a methanol solvate, 4·1.70
MeOH. 5 shows coordination isomerism compared to compounds
3 and 4 in that two of the carboxylate ligands in the metal coordi-
nation spheres on the end copper atoms have been replaced by
solvent methanol. The magnetic behavior of the di- and trinuclear
species has been probed and reveals that at low temperatures the
metal centers exhibit strong ferromagnetic coupling, which, on war-
ming, switches to simple paramagnetic behavior.
scopic and structural data for [NaCu(CF
Cu(C COO)(TEAH )] ·2H O (2), [Cu (CH
(HCOO) (TEAH ] (4) and [Cu (CF
OH) ][CF CO (5)
N(CH CH OH) ); [TEAH
3
COO)
COO)
OO)
triethanolamine
mono-deprotonated ion;
3
(TEAH
(TEAH
(TEAH
3
)] (1),
[
6
H
5
2
2
2
3
3
4
2 2
) ]
) -
2 2
(
(
(
3), [Cu
3
4
2
)
2
3
3
2
CH
3
2
3
2
]
2
(TEAH
3
ϭ
Ϫ
2
2
3
2
]
ϭ
Ϫ
[
TEAH]2 ϭ doubly deprotonated ion). Two approaches to these
compounds were employed:direct reaction of copper(II) carboxyla-
tes with TEAH and reaction of the tetrameric compound
{Cu(TEAH)} ] (6) with copper(II) carboxylates. Compound 1 is
3
[
4
Keywords: Copper; Copper triethanolamine-carboxylate complexes
Introduction
triethanolamine, N(CH CH OH) , complexes of copper
and bismuth for their potential application in the pro-
2
2
3
In the past decade there has been an increasing interest in
metal alkoxides because of their application in the sol-gel
synthesis of the various homo- and heterometallic oxide
materials. Metal alkoxides are easily hydrolyzed or pyro-
lyzed to give highly pure and homogeneous products, often
under far milder conditions than normally possible with
conventional solid state methods [1Ϫ6]. Using “soft” meth-
ods such as this allows better control of the fine structure
of the products, and makes possible controlled production
of nanoparticle metal oxide materials with narrow polydis-
persity under the proper conditions. Our work has emphas-
ized the formation of heterometallic complexes of bismuth
and other metals for applications as ferroelectric materials,
oxide ion conducting materials, and catalysts [7Ϫ12]. For
highly Lewis acidic metal centers such as bismuth, lead, and
the lanthanides, pre-formed metal complexes can simply be
attached to the metal center in a donor-acceptor fashion
duction of high T superconductors [13Ϫ15]. In this paper,
we will abbreviate neutral triethanolamine as TEAH3,
C
where H represents the active hydroxyl protons. Depro-
Ϫ
2
tonation can therefore lead to metal-bound TEAH
TEAH and TEA ions.
,
2Ϫ
3Ϫ
Our previous studies of copper complexes of triethanol-
amine reported the tetrameric species [{Cu(TEAH)}4]
which can be prepared in a variety of solvated forms [14].
This material makes a convenient starting point for the
preparation of other copper triethanolamine derivatives
that are reported herein. Previous studies of trinuclear cop-
per complexes found three types of molecular structures:
isosceles or equilateral triangular arrays of the copper ions,
and linearly arranged copper ions. Additionally, magnetic
studies have shown that these complexes may display either
ferromagnetic, anti-ferromagnetic, or both behaviors. Thus,
these exchange-coupled systems are important in the search
for molecular ferromagnets.
[
7Ϫ10]. In particular, we and others have had an interest in
In this paper we report the new copper triethanol-
amine
complexes
[NaCu(TEAH )(CF COO) ]
(1),
3
3
3
*
Prof. Dr. Kenton H. Whitmire
Department of Chemistry, MS-60, Rice University
100 Main St.,
[Cu(TEAH )(C H COO)] ·2H O (2), Cu (TEAH ) (CH
2
6
5
2
2
3
2 2
3-
COO)4
(3),
Cu (TEAH ) (HCOO)
4
(4)
and
3
2 2
6
Ϫ
[
Cu (TEAH ) (CF OO) (CH OH) ][CF CO ]2 (5) pre-
3 2 2 3 2 3 2 3 2
Houston TX, 77005 / USA,
pared by direct reaction of copper(II) carboxylates
e-mail: whitmir@rice.edu
Phone: 713-348-5650, Fax:713-348-5155
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
(RCO2 ϭ HCO2 , CH CO2 , CF CO2 , C H CO )
3 3 6 5 2
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876
DOI: 10.1002/zaac.200500204
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
2867

R. M. Escovar, J. H. Thurston, T. Ould-Ely, A. Kumar, K. H. Whitmire
with triethanolamine or reaction of the pre-formed tetra-
becoming slightly turbid. The reaction mixture was stirred at room
temperature for approximately 30 minutes during which time a pale
blue precipitate formed. The precipitate was removed by filtration
and the clear blue-green filtrate was concentrated to ca. one-quar-
ter of its original volume under reduced pressure. Storage of this
solution at reduced temperature resulted in the deposition of blue-
green crystals of 2. Yield: 78 %.
meric complex [{Cu(TEAH)} ] (6) with copper(II) car-
4
boxylates. The syntheses we describe are readily performed
in air and aqueous solutions from common off-the-shelf
reagents and produce materials free of excess TEAH3.
Experimental Section
Method 2: A suspension of benzoic acid (5 g, 0.04 mol) in deionized
water was treated with a solution of NaOH (1.64 g, 0.041 mol) in
deionized water with good stirring. The resulting white flocculent
precipitate was taken to dryness using a rotary evaporator. Half of
the resulting solid was suspended in deionized water and treated
with a solution of CuSO ·5H O (2.57 g, 0.010 mol) in deionized
water. The mixture was stirred and gently heated for approximately
15 minutes to produce a pale blue precipitate. The solid was col-
lected via filtration using a medium porosity Büchner funnel and
dried under reduced pressure.
General
The starting materials copper(II) acetate, copper(II) formate, ben-
zoic acid and triethanolamine were purchased and used as received.
All solvents employed in the reactions were purified by distillation.
Infrared spectra of the complexes were recorded of a Thermo-Nico-
let 630 instrument employing ATR techniques with a germanium
window. UV-Vis data of the complexes in water was collected on a
GBC spectral instrument. Mass spectra were obtained on a Finni-
gan MAT95 spectrometer using FAB with nitrobenzyl alcohol as
the matrix. Magnetic susceptibility data were measured on a Quan-
tum Design MPMS-5S SQUID magnetometer at 1000T from 2 to
4
2
A methanol solution of the copper(II) benzoate prepared as de-
scribed above was treated sequentially with a solution of benzoic
acid (0.57 g, 0.0050 mol) in methanol and an aqueous solution of
6 (0.994g, 0.0010 mol). The reaction mixture was stirred at room
temperature for approximately 15 minutes to give a blue-green solu-
tion and a pale blue precipitate. The supernatant was removed via
filtration and concentrated to ca. one-quarter of its original vol-
ume. Storage of the filtrate in a refrigerator resulted in the growth
of blue-green crystals that were found to be identical to those de-
scribed in Method 1 as determined by the crystallographic unit cell
parameters. Yield: 71 %. Elemental Analysis: %Found (%Calc.) for
[Cu (TEAH ) (C H COO) ]·2H O C, 44.33 (44.45); H, 6.15 (6.03);
300 K. Elemental analyses were obtained from Galbraith Analyti-
cal Laboratories.
Syntheses
[
NaCu(CF
Dissolution of Cu(SO)
5 mL) was accomplished with magnetic stirring, yielding a blue-
green solution. To this solution was added a solution of Na CO
0.454 g, 0.004 mol) in deionized water (ca. 25 mL). A light blue
3
CO
2
)
3
{TEAH
3
}] (1)
4
·5H
2
0 (1.0 g, 0.004 mol) in methanol (ca.
2
2
3
2
2 2
6
5
2
2
Ϫ1
(
N, 3.99 (3.99). FT-IR (cm ): 3103(O-H), 1597(COO)asym,
precipitate developed and was allowed to settle. Without any
further purification, trifluoroacetic acid (1.85 g, 0.020 mol) was ad-
ded and all the precipitate dissolved yielding a clear, pale blue solu-
1552(COO)sym, 1393, 1093, 1070, 1054, 1026, 907, 845, 813, 726,
Ϫ1
718. UV-Vis: λmax ϭ 267, 733 with ε ϭ 8350.8, 70.6 mol . The
Ϫ3
UV band, for concentrations up to 2.74 x 10 M followed Beer’s
tion. To this mixture was added [{Cu(TEAH)}
4
] (6) (0.947 g,
Law with a correlation coefficient of 0.968, while the visible band
Ϫ3
0.0010 mol) in deionized water (ca. 25 mL) yielding a precipitate.
for concentrations up to 5.4 ϫ 10 M followed Beer’s Law with a
2
The reaction was then filtered through a medium porosity glass frit
with the aid of Celite, and the filtrate was taken to dryness on a
rotary evaporator. The product was dissolved in warm methanol
R ϭ 0.993.
[Cu
3
(CH
3
COO)
4
(TEAH
2
)
2
] (3 and 3·2MeOH)
CCH (5.06 g, 0.025 mol) in
(ca. 50 mL) forming a cloudy solution. The precipitate was then
Method 1: To a solution of Cu(O
2
3 2
)
allowed to settle and was again filtered through a medium porosity
glass frit with the aid of Celite, yielding a blue-green solution. The
filtrate was then layered with diethyl ether and the product crys-
tallized over a period of several days. Yield: 17 %. Elemental Analy-
deionized water, triethanolamine (2.52 g, 0.017 mol) was added.
The solution immediately became deep blue. After stirring at room
temperature for about an hour, the solvent was removed under re-
duced pressure. The residue was dissolved in methanol and allowed
to recrystallize over a period of three days by slow evaporation.
The resulting solid was then redissolved in methanol and the re-
sulting bright blue solution was carefully layered with diethyl ether.
Solvent diffusion over the course of several days at room tempera-
ture resulted in the deposition of large blue prisms. After crystalli-
zation was judged to be complete, the solvent was decanted and
the product was washed with acetone (ca. 200 mL) then dried in
vacuo. Yield: 76 %, of a light blue powder or bright blue, clear
crystals. These crystals were found to be suitable for X-ray diffrac-
tion analyses; however, better crystals were obtained by slow vapor
diffusion under vacuum over a period of 4 weeks.
sis: %Found (% Calc.) for [NaCu(CF
3
CO
2
)
3
Ϫ1
(TEAH
3
)]: C, 25.12
(24.35), H, 3.11 (2.72), N, 2.92 (2.37). IR (cm ): 3401, 1675, 1433,
1
λ
202, 1185, 1129, 1092, 1074, 900, 846, 794, 727. UV-Vis:
Ϫ1
Ϫ1
max (ε) ϭ 253 nm (8658 M ), 742 nm (156 M ). The UV band,
Ϫ4
for concentrations up to 1 x 10 M, followed Beer’s Law with a
correlation coefficient of 0.933, while the visible band, for concen-
Ϫ3
2
trations up to 6.2 ϫ 10 M, followed Beer’s Law with a R
.9995.
ϭ
0
[Cu
2 6 5 2 2 2 2
(C H COO) (TEAH ) ]·2H O (2)
Method 1: A green, slightly turbid solution of CuSO
4
2
·5H O
(5.049 g, 0.020 mol) in methanol was treated with a solution of
Na CO (2.3 g, 0.022 mol) in deionized water resulting in the for-
2
3
Method 2: A solution of CuSO
ized water was treated with a solution of Na
0.021 mol) in deionized water, resulting in the immediate formation
of a pale blue precipitate. The resulting slurry, was then sequentially
treated with acetic acid (0.08 mol), which resulted in the formation
4
·5H
2
O (5.0 g, 0.020 mol) in deion-
CO (2.25 g,
mation of a pale blue precipitate. The slurry was then sequentially
treated first with a solution of benzoic acid (9.9 g, 0.081 mol) in
methanol, which resulted in the formation of a dark blue precipi-
tate, and then with a solution of 6 (8.96 g, 0.010 mol) suspended in
deionized water (ca. 100 mL). Addition of 6 resulted in the solution
2
3
of a clear blue solution; and then with [{Cu(TEAH)}
4
] (8.92 g,
2868
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876

New Mono-, Di-, and Trinuclear Copper(II) Triethanolamine-Carboxylate Complexes
.010 mol). The reaction mixture was stirred at room temperature of 0.933, while the visible band, for concentrations up to 6.2 x
0
Ϫ3
2
for 30 minutes, and then evaporated to dryness under reduced
pressure. Extraction of the residue with warm methanol resulted in
the formation of a cloudy blue-green solution, which was filtered
to remove a white solid. The clear filtrate was carefully layered with
diethyl ether, and solvent diffusion was allowed to occur over a
period of three days, during which time large, well-formed blue
crystals of 1 deposited in the flask. The product isolated in this
manner was found to give identical unit cell parameters as that
produced by Method 1. Yield: 72 %. Elemental Analysis: %Found
10 M, followed Beer’s Law with a R ϭ 0.9995.
[Cu
3
(CF
3
CO
2
)
2
(TEAH
2
)
2
(CH
3
OH)
2
][CF
3
CO
2
]
2
(5)
A deep turquoise solution of Cu(O
2
CCH
3 2
) (3.00 g, 0.015 mol) in
methanol was prepared with gentle heating and stirring. The solu-
tion was treated with trifluoroacetic acid (3.43 g, 0.030 mol), re-
sulting in a color change to pale blue. The mixture was stirred at
room temperature for 10 minutes and dried under reduced pressure.
The residue was dissolved in a minimal amount of methanol and
subsequently treated with triethanolamine (1.50 g, 0.010 mol),
which resulted in a color change to deep blue. The reaction mixture
was stirred at room temperature for 15 minutes and then again
evaporated to dryness under reduced pressure. The residue was ex-
tracted into a minimal amount of methanol and carefully layered
with diethyl ether. Solvent diffusion over a period of days at room
temperature resulted in the deposition of large, well-formed, blue-
green crystals. When crystallization was judged to be complete, the
solvent was decanted and the product was dried under reduced
pressure. Yield: 28 % of large, blue-green, clear crystals. The yield
(%Calc.) for Cu
3 2 2 3 4
(TEAH ) (CH COO) ·2MeOH: C, 33.48 (33.56);
Ϫ1
H, 6.20 (6.15); N, 3.64 (3.56) %. FT-IR (cm ): 2875(O-H),
1
1
568(COO)asym, 1443(COO)sym,1389, 1332, 1266, 1140, 1085, 1060,
026, 1002, 915, 902, 867, 750. UV-Vis: λmax ϭ 262 and 741 nm
Ϫ1
with ε ϭ 6038.4 and 189 mol , respectively. The UV band, for
concentrations up to 1.02 x 10 M, followed Beer’s Law with a
correlation coefficient of 0.963, while the visible band, for concen-
trations up to 5 ϫ 10 M , followed Beer’s Law with R ϭ 0.995.
Ϫ3
Ϫ3
2
[
Cu
This complex was produced in a manner analogous to 1, with the
substitution of Cu(O CH) for Cu(O CCH . Evaporation of the
3 4 2 2 3
(HCOO) (TEAH ) ]·2CH OH (4)
could be improved by allowing the product to crystallize over a
longer period of time. FT-IR (cm ): 3156(O-H), 1672(COO)asym
2
2
2
3
)
2
Ϫ1
,
reaction mixture under reduced pressure gave a blue solid, which
on recrystallization via diethyl ether/methanol diffusion gave the
product as a pale blue powder. Crystals suitable for x-ray diffrac-
tion studies were grown by further recrystallization of the product
by vapor phase diffusion of diethyl ether into a methanol solution
of the complex. Yield: 76 % of a light blue powder or bright blue
clear crystals. Elemental Analysis: %Found (%Calc.) For
1
1
7
567(COO)sym, 1483, 1439, 1354, 1203, 1132, 1086, 1066, 1046,
026, 1006, 904, 854, 796, 754, 733, 717. UV-Vis: λmax ϭ 267 nm,
Ϫ1
60 with ε ϭ 7743.3, 188.8 mol . The UV band for concentrations
Ϫ4
up to 1.44 x 10 M followed Beer’s Law with a correlation coef-
ficient of 0.997, while the visible band for concentrations up to 5.3
ϫ 10 M followed Beer’s Law with a R ϭ 0.999.
Ϫ3
2
Cu
N, 4.10(4.03). FT-IR (cm ): 3154, 1576, 1458, 1378, 1366, 1337,
086, 1050, 1024, 1006, 905, 867, 777, 750. UV-vis: λmax (ε) ϭ 253
3 2 2 4 3
(TEAH ) (HCOO) ·CH OH: C, 27.28 (31.10), H, 5.59(5.22),
Ϫ1
Single-Crystal X-ray Diffraction Studies
1
Ϫ1
Ϫ1
(
8658 M ), 742 (156 M ). The UV band, for concentrations up
Data for compounds 1, 3·2MeOH, 4·1.70MeOH and 5 were studied
on a Bruker Smart 1000 diffractometer equipped with a CCD area
Ϫ4
to 1 ϫ 10 M, followed Beer’s Law with a correlation coefficient
Table 1 Selective Crystallographic Data Collection and Refinement Parameters for Compounds 1 and 2
1
2
Composition
Chemical Formula
Formula wt /g·mol
Z
[NaCu(CF
3
CO
2
)
3
(TEAH
3
)]
ϱ
6 5 2 2 2 2
[Cu(C H CO )(TEAH )] ·2H O
C H42Cu N O
26 2 2 12
701.70
2
C
12
H15CuF
9
NNaO
9
(empirical)
Ϫ1
574.78
4
Cell dimensions:
˚
a /A
12.505(3)
10.834(2)
15.031(3)
98.69(3)
7.5920(15)
27.018(5)
8.0400(16)
111.05(3)
˚
b /A
˚
c /A
β /°
Cell Volume /A3
Temperature /K
Crystal system
Space Group
˚
2012.9(7)
293(2)
1539.2(5)
293(2)
monoclinic
P2 /n (#14)
1
1.514
monoclinic
P2
1
/n (#14)
Ϫ3
ρ /g·cm , calc
1.897
1148
F000
732
R
int
0.0427
0.0300
0 Յ h Յ 9
0 Յ k Յ 35
Ϫ10 Յ l Յ 9
3549
190
Data Collection Ranges
Ϫ13 Յ h Յ 7
Ϫ12 Յ k Յ 11
Ϫ16 Յ l Յ 16
2903
296
0.0949
Number of reflections
Number of Paramaters
R (all reflections)
0.1210
R (I Ն 2σ (I))
0.0726
0.2215
0.0601
0.2057
R
w
(all reflections)
Goodness of fit, S
Max shift/error
Residual peaks max/min /e/A
1.129
0.001
1.463/Ϫ0.843
1.132
0.000
1.246/Ϫ1.974
˚
3
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876
zaac.wiley-vch.de
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
2869

R. M. Escovar, J. H. Thurston, T. Ould-Ely, A. Kumar, K. H. Whitmire
Table 2 Selective Crystallographic Data Collection and Refinement Parameters the Trinuclear Copper Compounds 3, 3·2MeOH, 4, and 5
3
Cu
3·2 MeOH monoclinic
Cu (CH CO (TEAH
MeOH
48Cu
787.24
2
3·2 MeOH orthorhombic* 4·1.70 MeOH
· Cu (CH CO (TEAH · Cu (HCO (TEAH
2 MeOH 1.70 MeOH
48Cu
5
Composition
3
(CH
3
CO
2
)
4
(TEAH
2
)
2
3
3
2
)
4
2
)
2
3
3
2
)
4
2
)
2
3
2
)
4
2
)
2
· [Cu
3
(CH
3
CO
][CF
2
)
3
2
(TEAH
2 2
) -
2
C
(MeOH)
22HCu
2
2 2
CO ]
Chemical Formula
Formula wt /g·mol
Z
C
723.16
1
20
H40Cu
3
N
2
O
14
22
H
3
N
2
O
16
C
22
H
3
N
2
O
16
C
17.0
H
38.80Cu
3
N
2
O
15.70
C
3
N
2
O
16
Ϫ1
787.24
4
721.53
1
997.10
1
Cell dimensions:
˚
a /A
7.7550(16)
9.791(2)
10.235(2)
69.74(3)
82.81(3)
78.36(3)
712.8(2)
293(2)
7.8969(17)
10.599(2)
10.732(4)
19.545(4)
7.9530(16)
20.724(4)
7.9418(16)
8.6198(17)
11.412(2)
72.15(3)
85.02(3)
69.33(3)
695.6(2)
293(2)
8.4906(17)
10.637(2)
11.035(2)
112.51(3)
98.95(3)
98.24(3)
886.8(3)
293(2)
˚
b /A
˚
c /A
Ͱ /°
β /°
γ /°
92.216(4)
˚
3
Cell Volume /A
1650.2(6)
293(2)
3222.9(11)
293(2)
orthorhombic
Pbca
Temperature /K
Crystal system
Space Group
triclinic
monoclinic
P2
triclinic
triclinic
¯
¯
¯
P1(#2)
1
/n(#14)
P1(#2)
P1(#2)
Ϫ3
ρ /g·cm , calc
1.685
373
1.584
1.722
1.867
499
F000
818
372
R
int
0.012
0.0669
Ϫ8 Յ h Յ 7
Ϫ11 Յ k Յ 9
0.0196
Ϫ8 Յ h Յ 8
Ϫ8 Յ k Յ 9
0.0119
Ϫ9 Յ h Յ 9
Ϫ7 Յ k Յ 11
Data Collection Ranges Ϫ8 Յ h Յ 8
Ϫ9 Յ k Յ 0
Ϫ9 Յ l Յ 10
1353
179
0.0582
0.0506
Ϫ21 Յ l Յ 19
2373
196
0.0598
0.0496
Ϫ12 Յ l Յ 12
1942
191
0.0555
0.0457
Ϫ12 Յ l Յ 11
2530
238
0.0742
0.0713
#
#
Reflections
Paramaters
R (all reflections)
R (I Ն 2σ (I))
R
w
(all reflections)
0.1982
1.228
0.1528
1.053
0.1306
1.055
0.2155
1.111
Goodness of fit, S
Max shift/error
peaks max/min /e/A
0.000
0.744/Ϫ0.821
0.001
0.696/Ϫ0.498
0.000
1.033/Ϫ0.637
0.000
1.693/Ϫ1.435
˚
3
*
This data set refined poorly but sufficiently to establish the identity of the compound, thus only minimal data are provided.
detector. Data for compounds 2 and unsolvated 3 were collected
on a Rigaku AFC5S diffractometer. In all cases, graphite-monoch-
˚
Table 3 Selected Bond Distances/A and Angles/° for Compound 1.
romated MoK
α
radiation was employed. The stability of the crys-
Na(1)-O(11)#2
Na(1)-O(21)
Na(1)-O(31)
Na(1)-O(42)
Na(1)-O(52)#1
Na(1)-O(61)
2.607(6)
2.717(6)
2.570(6)
2.374(6)
2.404(7)
2.298(6)
Cu(1)-N(1)
Cu(1)-O(11)
Cu(1)-O(21)
Cu(1)-O(31)
Cu(1)-O(41)
Cu(1)-O(51)
2.032(6)
2.642(6)
1.967(5)
2.391(5)
1.942(5)
1.936(6)
tals during data collections was monitored through three standard
reflections which were collected every 75 frames on the Bruker
Smart 1000 while standards were measured for intensity variations
after every 150 reflections measured on the Rigaku AFC5S. No
significant decay of the reflection intensities were detected during
the data collections for any of the crystals. Pertinent details of the
data collections and refinements are supplied in Tables 1 and 2.
The data were corrected for Lorentz and polarization effects. Ab-
sorption corrections for the data measured on the Bruker Smart
O(51)-Cu(1)-O(41) 90.7(2)
O(51)-Cu(1)-O(21) 178.0(2)
O(41)-Cu(1)-O(21) 90.5(2)
O(51)-Cu(1)-N(1)
O(41)-Cu(1)-N(1) 171.0(2)
O(21)-Cu(1)-N(1) 84.8(2)
O(51)-Cu(1)-O(31) 93.3(2)
O(41)-Cu(1)-O(31) 109.3(2)
O(21)-Cu(1)-O(31) 87.8(2)
O(61)-Na(1)-O(42)
O(61)-Na(1)-O(52)#1
O(42)-Na(1)-O(52)#1
O(61)-Na(1)-O(31)
O(42)-Na(1)-O(31)
O(52)#1-Na(1)-O(31)
O(61)-Na(1)-O(11)#2
O(42)-Na(1)-O(11)#2
162.4(3)
111.9(2)
83.9(2)
87.8(2)
79.1(2)
110.5(2)
93.8(2)
89.9(2)
93.8(2)
1000 system were applied using the program SADABS [16]. The
data for compounds 2 and unsolvated 3 that were collected on a
Rigaku AFC5 diffractometer were corrected for absorption using
psi scans. Data collection and reduction for the AFC5S data were
performed using TEXSAN v. 5.0 [17], while those collected on the
Bruker system were performed using the SHELXTL package [18].
Pertinent details of the data collections are supplied in Tables 1
O(52)#1-Na(1)-O(11)#2 102.6(2)
N(1)-Cu(1)-O(31)
78.3(2)
O(31)-Na(1)-O(11)#2
O(61)-Na(1)-O(21)
O(42)-Na(1)-O(21)
O(52)#1-Na(1)-O(21)
O(31)-Na(1)-O(21)
O(11)#2-Na(1)-O(21)
143.6(2)
90.1(2)
74.47(19)
158.0(2)
70.09(18)
73.54(18)
O(51)-Cu(1)-O(11) 87.8(2)
O(41)-Cu(1)-O(11) 97.1(2)
O(21)-Cu(1)-O(11) 90.5(2)
N(1)-Cu(1)-O(11)
O(31)-Cu(1)-O(11) 153.55(18)
75.2(2)
(compounds 1 and 2) and 2 (compounds 3 Ϫ 5). For 4, the lattice
methanol was found in the electron density difference maps and
refined at partial occupancy to give an over formula of
Symmetry transformations used to generate equivalent atoms: #1 Ϫx, Ϫyϩ3,
Ϫzϩ2 #2 Ϫxϩ1/2, yϩ1/2, Ϫzϩ5/2 #3 Ϫxϩ1/2, yϪ1/2, Ϫzϩ5/2
4·1.70MeOH.
In addition to the unsolvated form of compound 3, methanol sol-
vated forms in both monoclinic and orthorhombic settings were
observed. The latter orthorhombic setting was solved sufficiently
to confirm the composition as being the same as for the monoclinic
form, but the data were not as good and a number of shadow peaks
remained near the heavy atoms in the final refinements possibly
indicating a less-than-optimal absorption correction or some irre-
solvable disorder. Since the monoclinic form refined well, it was
not deemed worthwhile to pursue further study on the orthorhom-
bic form; however, the crystal parameters of that form are pre-
sented in Table 2 for comparison purposes. Plots of the structures
are given in Figures 1Ϫ9. For all thermal displacement parameter
plots, the ellipsoids are drawn at the 50 % probability level.
Heavy atoms in all six compounds were located using direct meth-
ods with the SHELXTL software package [18]. All other atoms
were located by successive Fourier difference maps and were refined
2870
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876

New Mono-, Di-, and Trinuclear Copper(II) Triethanolamine-Carboxylate Complexes
2
˚
using the full-matrix least squares technique on F . Hydrogen
atoms were placed in calculated positions and allowed to ride on
the adjacent atom. Refinement of positional and anisotropic dis-
placement parameters for all non-hydrogen atoms (except for those
described below for 1 and 5) led to convergence in all cases. A list
of relevant bond lengths and angles of the molecules are supplied
in Tables 3Ϫ5.
Table 4 Selected Bond Distances/A and Angles/° for Compound
2.
Cu(1)-N(1)
Cu(1)-O(11)
Cu(1)-O(21)
2.053(4)
2.474(5)
1.951(3)
Cu(1)-O(21)#1
Cu(1)-O(31)
Cu(1)-O(1)
1.946(3)
2.570(4)
1.966(4)
O(21)#1-Cu(1)-O(21) 82.69(16)
O(1)-Cu(1)-O(11)
N(1)-Cu(1)-O(11)
O(21)#1-Cu(1)-O(31) 114.71(15)
O(21)-Cu(1)-O(31)
O(1)-Cu(1)-O(31)
N(1)-Cu(1)-O(31)
O(11)-Cu(1)-O(31)
85.71(16)
76.38(16)
O(21)#1-Cu(1)-O(1)
95.16(16)
O(21)-Cu(1)-O(1)
177.28(15)
3
For 1 and 5, the CF groups were found to be disordered. Aniso-
tropic refinement of a single position, or even two disordered posi-
tions, led to unsatisfactory thermal displacement ellipsoids even
O(21)#1-Cu(1)-N(1) 161.62(17)
90.71(15)
88.68(15)
78.66(15)
153.39(13)
O(21)-Cu(1)-N(1)
O(1)-Cu(1)-N(1)
84.66(16)
97.82(17)
when the CF
ideal tetrahedral geometries at the C atoms. For 1, one CF
was found to be ordered, while the other two were refined with
three orientations of idealized CF groups (total population of 1).
3
groups were refined with restraints to give nearly
O(21)#1-Cu(1)-O(11) 91.72(15)
O(21)-Cu(1)-O(11) 95.99(16)
3
group
Symmetry transformation used to generate equivalent atoms: #1 Ϫxϩ1,
Ϫy, Ϫzϩ2
3
During the final stages of refinement, these populations were fixed
and a common isotropic thermal parameter for all fluorine atoms
3
was refined. For 5, two orientations of each CF group were em-
ployed with total occupancies summing to 1, and with a common
isotropic thermal parameter for each fluorine atom. The formate
˚
Table 5 Comparison of Bond Distances/A and Angles/° for the
Trinuclear Compounds 3, 3·2 MeOH, 4 and 5
derivative 4 also exhibits disorder in one of the formate ligands.
Ϫ
The terminal HCO
2
group shows two sites with roughly equal
Distances to the central copper atoms:
populations for the non-metal-bound O atom.
3
3·2 MeOH 4·1.70 MeOH 5
monoclinic
Results and Discussion
Cu(1)-O(21)
Cu(1)-O(31)
Cu(1)-O(41)
1.929(6) 1.915(3)
2.730(7) 2.764(4)
1.950(7) 1.961(4)
1.964(3)
2.523(4)
1.969(4)
1.956(4)
2.557(6)
1.976(5)
Syntheses
The triethanolamine ligand readily binds to the copper(II)
ion [14], and in the absence of coordinating ligands such as
Distances to the end copper atoms:
carboxylate ions produces tetrameric [{Cu(TEAH)} ] (6).
In pursuing other synthetic routes to 6, we attempted to
Cu(2)-O(11)
Cu(2)-O(21)
Cu(2)-O(31)
Cu(2)-O(42)
Cu(2)-O(51)
Cu(2)-N(1)
2.508(7) 2.563(4)
2.429(4)
1.950(4)
2.528(4)
1.953(4)
1.958(4)
2.024(4)
2.391(5)
1.950(5)
2.437(5)
1.951(5)
2.008(5)
2.035(6)
4
1.916(6) 1.924(3)
2.535(7) 2.467(3)
1.945(7) 1.958(4)
1.969(7) 1.987(3)
2.058(8) 2.046(4)
treat copper(II) acetate directly with TEAH , with the idea
3
that removal of volatile acetic acid would drive the reaction
to completion. Upon heating, solutions of copper(II) acet-
ate and TEAH3 emitted a strong smell of acetic acid.
Characterization of crystals from the first attempt at this
reaction showed, that the acetate ligands were not com-
pletely replaced. Instead mixed complexes containing both
triethanolamine residues and carboxylate were obtained.
Bond angles involving the central copper atoms:
O(21)-Cu(1)-O(21)#Symm 180.0(3) 180.000(1) 180.000(1)
180.000(1)
81.63(18)
98.37(18)
90.9(2)
89.1(2)
88.7(2)
O(21)-Cu(1)-O(31)
O(21)#Symm-Cu(1)-O(31) 101.1(2) 101.94(12)
O(21)-Cu(1)-O(41) 89.6(3) 89.52(15)
O(21)-Cu(1)-O(41)#Symm 90.4(3) 90.48(15)
O(31)-Cu(1)-O(41) 93.9(3) 92.25(14)
78.9(2) 78.06(12)
83.54(13)
96.46(13)
88.75(15)
91.25(15)
91.30(16)
88.70(16)
180.000(1)
Based upon these observations, a variety of heteroleptic
Ϫ
copper(II) carboxylate complexes of [TEAH2]
or
O(31)-Cu(1)-O(41)#Symm 86.1(3) 87.75(14)
O(41)-Cu(1)-O(41)#Symm 180.0(3) 180.0(2)
91.3(2)
180.000(1)
[TEAH]2 were prepared via syntheses in aqueous or meth-
Ϫ
anolic solutions by either the addition of TEAH to a cop-
per(II) carboxylate solution or the addition of a copper(II)
3
Bond angles involving the end copper atoms:
N(1)-Cu(2)-O(11)
N(1)-Cu(2)-O(21)
N(1)-Cu(2)-O(31)
N(1)-Cu(2)-O(42)
N(1)-Cu(2)-O(51)
O(11)-Cu(2)-O(21)
O(11)-Cu(2)-O(31)
O(11)-Cu(2)-O(42)
O(11)-Cu(2)-O(51)
O(21)-Cu(2)-O(31)
O(21)-Cu(2)-O(42)
O(21)-Cu(2)-O(51)
O(31)-Cu(2)-O(42)
O(31)-Cu(2)-O(51)
O(42)-Cu(2)-O(51)
76.5(3) 76.04(15)
85.8(3) 85.40(15)
76.1(3) 77.79(14)
177.7(3) 178.16(14) 177.24(16)
94.2(3) 94.48(16)
92.8(2) 91.86(14)
152.6(2) 153.82(14) 153.95(13)
102.0(3) 105.66(15) 103.28(15)
92.5(2) 92.09(14)
84.3(2) 85.94(13)
92.6(3) 93.80(15)
174.5(3) 175.90(13) 176.57(16)
105.4(3) 100.51(14) 102.75(15)
90.3(2) 90.03(13)
87.5(3) 86.20(15)
76.81(15)
86.65(16)
77.14(15)
77.5(2)
86.7(2)
78.6(2)
177.8(2)
92.9(2)
97.44(19)
155.80(19)
104.6(2)
85.7(2)
84.98(19)
92.4(2)
176.64(19)
99.3(2)
91.7(2)
87.8(2)
carboxylate to 6. In the first method, the reaction is
an equilibrium between free TEAH3 and metal-bound
xϪ
carboxylate, versus metal-bound [TEAH3Ϫx
]
and free
93.19(18)
95.02(14)
carboxylic acid. This equilibrium can be shifted by the
removal of the volatile carboxylic acid product (Equa-
tion (1)).
88.27(17)
83.68(14)
90.60(15)
3 2 2 2 2 2
TEAH ϩ Cu(O CR) Ǟ [Cu(O CR) (TEAH )] ϩ RCO HȆ (1)
Because this approach depends upon the removal of the
carboxylic acid, it does not work well for the heavier carb-
oxylates due to their lower volatility, and therefore a differ-
ent route to synthesize complexes of the less volatile carb-
oxylic acids was sought. This was accomplished by the stoi-
92.94(17)
89.58(17)
Symmetry transformations used to generate equivalent atoms for compound
3
: #Symm ϭ ϪxϪ1, Ϫyϩ1, Ϫz; Compound 3·2 MeOH monoclinic form:
#
Symm ϭ Ϫx, Ϫyϩ2, Ϫzϩ1; Compound 4: Symm ϭ ϪxϪ1, Ϫyϩ2, Ϫzϩ1;
Compound 5: #Symm ϭ Ϫxϩ1, Ϫyϩ1, Ϫzϩ1
chiometric addition of [{Cu(TEAH)} ] to an in situ-pre-
4
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876
zaac.wiley-vch.de
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
2871

R. M. Escovar, J. H. Thurston, T. Ould-Ely, A. Kumar, K. H. Whitmire
pared copper(II) carboxylate. The latter was synthesized by
ease of discussion, the three acetate units will be described
as trifluoroacetate(x), where x ϭ 4, 5, or 6, corresponding
to the atom labels C(x1), C(x2), O(x1) and O(x2). This com-
pound is polymeric in the solid state forming an infinite
2-dimensional sheet (Figure 2). Trifluoroacetate(6) is bound
in a terminal fashion only to the sodium ion, while tri-
first producing a slurry of ’Cu(OH) ’ via the reaction of
2
aqueous copper(II) sulfate and sodium carbonate followed
by treatment with an appropriate carboxylic acid. In the
case of trifluoroacetic acid, if residual sodium ions from the
sodium carbonate employed are not removed, a double salt
of sodium trifluoroacetate and copper(II) acetate coordi-
ϩ
fluoroacetates(4) and (5) bridge between the Na ions and
the Cu2 centers. Two oxygen atoms (O(21) and O(31)) of
ϩ
nated by TEAH is obtained.
3
The formulations of the compounds observed show dif-
ferent degrees of displacement of the anionic carboxylate
ligands by a triethanolamine-derived ligand. For Com-
the TEAH ligand on one copper center also bridge to one
sodium ion, while O(11) bridges to a different Na ion.
3
ϩ
pounds 1 and 2 the ratio of TEAH to Cu is 1:1. Com-
3
pound 1 shows a neutral, non-deprotonated form of the
ϩ
TEAH ligand, based upon the charge balance (one Na
3
2
ϩ
Ϫ
ion, one Cu ion and three CF CO ions). This simple
3
2
coordination of TEAH is unique for this series of com-
3
pounds as the other compounds show either single or
Figure 2 A partial packing diagram of 1 showing the 2-dimensio-
nal sheet-like structure.
double deprotonation of TEAH upon complexation. The
3
dinuclear benzoate derivative is also a 1:1 complex between
copper and a TEAH -derived ligand, but here the single
3
deprotonation of TEAH allows for the elimination of one
3
carboxylate ligand producing [Cu(C H CO )(TEAH )] (2).
6
5
2
2 2
The known tetranuclear complex 6 is also a 1:1 complex,
but here the amine ligand is doubly deprotonated so that
the copper valence is completely satisfied by the chelating
ligand and no other ligands are present. For the trinuclear
Ϫ
species 3, 4, and 5, the stoichiometry is 3:2 (Cu: TEAH2 ),
so the complexes may be viewed as triethanolamine poor
relative to 1 and 2. One way of viewing these compositions
Ϫ
is as 2:1 adducts of TEAH2 /carboxylate complexes anal-
ogous to 2 with an additional equivalent of copper(II) car-
boxylate:[Cu(O CR)(TEAH )] [Cu(O CR) ].
2
2
2
2
2
Structures
Compound 1 is a double salt of sodium trifluoroactetate
and copper(II) trifluoroacetate, with a neutral triethanol-
amine ligand, TEAH , coordinated to the copper atom. The
3
asymmetric unit of the structure is shown in Figure 1. The
overall composition is [NaCu(CF CO ) (TEAH )] . For
ϩ
Figure 3 A ball and stick plot showing the Cu2 coordination
environment in 1.
3
2 3
3 ϱ
ϩ
Figure 4 A ball and stick plot showing the Na coordination en-
Figure 1 The asymmetric unit for compound 1.
872
vironment in 1.
2
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876

New Mono-, Di-, and Trinuclear Copper(II) Triethanolamine-Carboxylate Complexes
Overall, this provides distorted octahedral coordination for
both Cu (Figure 3) and Na ions (Figure 4). The Cu
ion is thus coordinated by the nitrogen and three oxygen
The linear trinuclear motif appears to be a favored
2
ϩ
ϩ
2ϩ
arrangement, and such compositions have been observed
for the acetate, formate, and trifluoroacetate ligands. The
core structure has been observed previously for related
complexes based upon N,N-2-diethylaminoethanol:
[Cu (RCO ) (Et NCH CH O) ] although those com-
atoms of the TEAH ligand as well as two carboxylate oxy-
3
xϪ
gen atoms. As with all of the copper TEAH3Ϫx (x ϭ
0
Ϫ 2) complexes observed to date, the N(CH CH O) unit
2 2 3
3
2 4
2
2
2
2
coordinates the metal with the oxygen atoms in a
pounds were found to have C symmetry rather than i as
2
meridional configuration, I. The Cu-OTEAH3 distances, as
found here [19Ϫ24]. All of these compounds have very simi-
with other TEAH derivatives such as 6 are highly asym-
3
˚
2ϩ
metric, ranging from 1.967(5) to 2.642(6) A [14]. The Cu
-
O
distances are shorter at 1.936(6) and
trifluoroacetate
˚
ϩ
1.942(5) A. The Na -O
distances range from 2.570(6)
TEAH
3
˚
ϩ
to 2.717(6) A, while the Na -O
shorter, lying between 2.298(6) and 2.404(7) A. As seen in
Table 3, the angles about Na and Cu are also highly
distorted from the idealized 90 and 180° values expected for
a regular octahedron.
distances are
trifluoroacetate
˚
ϩ
2ϩ
Figure 6 Thermal ellipsoid of unsolvated compound 3. Hydrogen
atoms have been omitted for clarity.
The benzoate derivative 2 can be viewed as a dimer of
two neutral Cu(TEAH )(C H CO ) units. The dimerization
2
6
5
2
is achieved through the bridging of one oxygen atom of
Ϫ
2ϩ
each TEAH2 unit to the adjacent Cu ion, and the mol-
ecule possesses crystallographic inversion symmetry. The
carboxylate groups are bound in a terminal fashion to each
copper atom. As indicated in Table 4 and Figure 5, the Cu-
OTEAH3 distances are highly asymmetric (Table 4) ranging
˚
from 1.951(3) to 2.570(4) A. With all interactions to copper
included, the metal center can be viewed as a highly dis-
torted octahedron similar to 1.
Figure 7 Thermal ellipsoid of monoclinic version of methanol sol-
vated compound 3·2MeOH. Hydrogen atoms have been omitted
for clarity.
Figure 5 Thermal ellipsoid plot of compound 2. Hydrogen atoms
have been omitted for clarity.
Figure 8 Thermal ellipsoid of the formate derivative compound
4. Hydrogen atoms have been omitted for clarity.
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876
zaac.wiley-vch.de
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
2873

R. M. Escovar, J. H. Thurston, T. Ould-Ely, A. Kumar, K. H. Whitmire
(
O C H Me-2) ](H O) [19] (II) in which the central copper
2 6 4 4 2 2
atom is ligated by two water molecules in addition to the
two carboxylate and two ethanolamine oxygen atoms.
The end copper atoms are also roughly octahedral but
there is a substantial difference between the acetate and for-
mate complexes as compared to the trifluoroacetate deriva-
tive. For both acetate and formate, the end copper atoms
have their coordination spheres completed by bonding to
Ϫ
the four donor atoms of the TEAH2 ligands, a carboxylate
ligand, which bridges to the central copper center and a
terminal carboxylate. Both of these molecules show crystal
morphologies that include in the lattice two molecules
(when considered at full occupancy) of methanol per tri-
nuclear copper complex. In the case of the trifluoroacetate
derivative, however, the end copper atoms are not bound
to a terminal carboxylate ligand. Instead, these atoms are
attached to methanol solvent and the additional trifluoro-
acetate ions are located in the crystal lattice. One can
consider these as examples of coordination iso-
merism: [Cu (RCO ) (TEAH ) (solvent) ][RCO ] versus
Figure 9 Thermal ellipsoid plot of the trinuclear trifluoroacetate
derivative 5. Hydrogen atoms have been omitted for clarity.
3
2 2
2 2
2
2 2
lar structural features and trends in bond distances and
angles. A comparison of all the compounds is found in
Table 5. The compounds all possess a central Cu atom that
is bridged to two other Cu atoms through carboxylate ions
[Cu (RCO ) (TEAH ) ]·2(solvent). Given the ability of tri-
3 2 4 2 2
ethanolamine to exist with several degrees of protonation
coupled with the coordination isomerism, there could be
some ambiguity about whether the CF CO units in the
3
2
Ϫ
Ϫ
as well as via two oxygen atoms of the TEAH2 ligand
lattice are present as CF CO2 or CF CO H given the diffi-
3 3 2
attached to the end copper atoms, with distances to O(31)
culty of identifying hydrogen atoms in the X-ray structure;
however, it appears unlikely that a relatively strong acid
such as trifluoroacetic acid would exist simultaneously with
a metal bound-alkoxide ligand, as would be required if the
CF CO H were the species present in the lattice.
are very long in each case (Figures 6 Ϫ 9), although these
˚
distances are very much longer (ϳ0.2 A) for the acetate
structures as compared to both the formate and trifluoro-
acetate structures. This may arise from the weaker acidity
of acetic acid (pK ϭ 4.75) than for formic acid (pK ϭ
3
2
a
a
3
.75) and trifluoroacetic acid (pK ϭ 0), which in turn is
a
related to the electron donating/withdrawing capability of
Magnetic Data for Compound 2
R in the RCO H. One might expect even stronger bonding
2
The temperature dependence of the magnetic susceptibility
of 2 in the temperature range 2-300 K is shown in Fig-
ure 10. As expected, the magnetic behavior of 2 resembles
that of related dinuclear compounds. At room temperature
of the triethanolamine oxygen atoms to copper for the
trifluoroacetate derivative as compared to the formate
compound; however, compound 5 differs in the number of
carboxylate ions bound directly to the metals (vide infra)
making comparisons less clear. As with 1 and 2 the central
copper atoms Cu(1) in all compounds can be viewed as
highly distorted octahedral centers considering the binding
of O(31), or as square planar copper centers if those long
interactions are neglected. This situation is similar to that
encountered for the long interactions as in the related tri-
nuclear ethanolamine complexes [Cu (OCH CH NEt ) -
3
Ϫ1
χ T is 0.91 cm ·mol ·K. This value is slightly larger than
M
II
that expected for a magnetically uncoupled dinuclear Cu
compound (0.82 cm ·mol ·K, given a g ϭ 2.1). Upon
3
Ϫ1
cooling the sample, χ T increases and reaches a maximum
M
Ϫ1
3
at 8 K of 1.3 cm ·mol ·K. This behavior is characteristic
of strong ferromagnetic exchange coupling between the ad-
3
2
2
2 2
2
Figure 10 Magnetic data for Compound 2·2H O showing a fitted
plot of µeff versus T(K).
2874
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876

New Mono-, Di-, and Trinuclear Copper(II) Triethanolamine-Carboxylate Complexes
jacent CuII atoms [19, 25]. Below this maximum the χ T
product decreases rapidly down to 1.09 cm ·mol ·K.
The appropriate Heisenberg-Hamiltonian for two para-
magnetic centers is the simple isotropic:
M
Table
4·1.70MeOH, and 5.
6 Magnetic Parameters for Compounds 3·2MeOH,
3
Ϫ1
Compound
J
12 /cmϪ1
g
θ /K
R
1
4
5
65(2)
88.9(3.6)
70.8 (1.5)
2.21(0.07)
2.07(0.01)
2.25 (0.003)
Ϫ0.6
Ϫ1.4
Ϫ0.2
8.2 E-6
1.7 E-6
4.6 E-7
ˆ
ˆ ˆ
H ϭ ϪJS S
1 2
where S ϭ S ϭ 1/2, and zero field splitting (ZFS) effects
1
2
have again been ignored [25]. The exchange parameter J
refers to the interaction between both copper centers. The
is caused by ZFS. Fitting the experimental data with equa-
ϩ
resulting spin levels ΗST ϭ S ϩS , S ϭ S ϩS ͘ for two
1
2
1
2
tion (2) using MAGMUN afforded the J , J , g, TIP, and
12
13
interacting S ϭ 1/2 centers gives two doublets Η1, 1͘ and
i
θ parameters [27]. These parameters and R (where R is cal-
Η0, 0͘. From this Hamiltonian the following version of the
Bleaney-Bowers expression for the temperature dependence
of χ is derived:
culated from the expression {Σ [(χ T)
R
ϭ
-
i
M
obs,i
2
2
(
χ T)
] / Σ [(χ T) ] }) are summarized in Table 6
calcd,i i M obs,i
M
for compounds 3, 4, and 5[25].
2
g2
Ϫ1
2
N
A
µ
B
1
3
Ϫ2J
χcalcd (T) ϭ
΄1 ϩ
exp
΂ ΃΅
3
k(TϪθ)
kT
where θ, the Weiss constant, is included to account for the
decrease in magnetic moments at low temperatures which
is caused by ZFS [19, 26]. Fitting the experimental data
with equation (4) using MAGMUN afforded J ϭ 70.5
Ϫ1
(
4.6) cm , g ϭ 2.11 (0.006), TIP ϭ 0, θ ϭ Ϫ0.02, and
R ϭ 1.6 E Ϫ7 (where R is calculated from the expression
2
2
R ϭ {Σ [(χ T) -(χ T)
] /Σ [(χ T) ] }) [25, 27].
i
M
obs,i
M
calcd,i
i M obs,i
The g tensor values evaluated from the fits are comparable
to those found in similar dinuclear compounds, and the
coupling constant J is indicative of strong ferromagnetic
coupling, albeit at low temperatures.
Figure 11 Magnetic data for Compound 3·2MeOH showing a
fitted plot of µeff versus T(K).
Magnetic Data for Compounds 3, 4 and 5
The appropriate isotropic Heisenberg-Hamiltonian for
three paramagnetic centers in C2v symmetry has the form
which should also appropriate for the inversion-symmetric
compounds reported here [19]:
ˆ
ˆ ˆ
ˆ ˆ
ϩ S
ˆ ˆ
1 3
) Ϫ 2J13(S S ),
H ϭ Ϫ2J12(S
1
S
2
2
S
3
where the occurrence of axial ZFS between the Kramer
doublets has not been included [25]. The exchange para-
meters J12 and J13 refer to the interaction between the two
adjacent centers and the interactions between the terminal
Figure 12 Magnetic data for Compound 4·1.70MeOH showing a
fitted plot of µeff versus T(K).
centers, respectively. The resulting spin levels ΗS
ϭ
T
ϩ
S ϩS ϩS , S ϭ S ϩS ͘ for three interacting S ϭ 1/2 cen-
1
2
3
1
3
i
ters are one quartet Η3/2, 1/2͘, and two doublets,Η1/2, 0͘ and
Η1/2, 1͘, with energies of E(3/2,1/2) ϭ -J -1/2J , E(1/2,0) ϭ
1
2
13
3
/2J , and E(1/2,1) ϭ 2J -1/2J , respectively. Applying
13 12 13
this to the van Vleck equation, the following theoretical ex-
pression for the temperature dependence of χ is obtained
[
19]:
N
A
g2 µ
B
2
10exp(J12 /kT) ϩ exp(Ϫ2J12 /kT) ϩ exp(Ϫ2J13 /kT)
χcalcd (T) ϭ
·
ϩ Nα
1
2k(TϪθ) 2exp(J12 /kT) ϩ exp(Ϫ2J12 /kT) ϩ exp(Ϫ2J13 /kT)
where θ, the Weiss constant, is included to account for the
decrease in magnetic moments at low temperatures which
Figure 13 Magnetic data for Compound 5 showing a fitted plot
of µeff versus T(K).
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876
zaac.wiley-vch.de
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
2875

R. M. Escovar, J. H. Thurston, T. Ould-Ely, A. Kumar, K. H. Whitmire
Fits were conducted both by fixing the terminal coupling
parameter at J13 ϭ 0 and by allowing it to vary freely. In the
latter case the free fit parameter was insignificantly small,
[2] R. C. Mehrotra, A. Singh, Prog. Inorg. Chem. 1997, 46, 239.
[
[
3] R. C. Mehrotra, A. Singh, Chem. Soc. Rev. 1996, 25, 1.
4] R. C. Mehrotra, J. Ind. Chem. Soc. 1994, 71, 315.
Ϫ1
[5] R. C. Mehrotra, A. Singh, S. Sogani, Chem. Rev. 1994, 94,
643.
converging to J13 ഠ 0 cm , and did not improve the qual-
1
ity of the fits. The g tensor values evaluated from the fits
are comparable to those found in similar trimeric com-
pounds [25]. The J12 exchange parameters in Table 6 are
indicative of strong ferromagnetic coupling, albeit at low
temperatures [19, 24, 25].
[6] R. C. Mehrotra, J. Non-Cryst. Solids 1990, 121, 1.
[7] J. H. Thurston, K. H. Whitmire, Inorg. Chem. 2002, 41, 4194.
[8] J. H. Thurston, K. H. Whitmire, Inorg. Chem. 2003, 42, 2014.
[9] J. H. Thurston, C. G.-Z. Tang, D. W. Trahan, K. H. Whitmire,
Inorg. Chem. 2004, 43, 2708.
The temperature dependence of the molar magnetic sus-
[10] J. Thurston, H. D. Trahan, T. Ould-Ely, K. H. Whitmire,
ceptibility χ T of 3, 4, and 5 in the temperature range
Inorg. Chem. 2004, 43, 3299.
M
[
[
[
[
11] J. H. Thurston, A. Kumar, C. Hofmann, K. H. Whitmire,
2
-300 K are shown in Figures 11Ϫ13. As expected, the mag-
Inorg. Chem. 2004, 43, 8427.
12] J. H. Thurston, T. Ould Ely, D. Trahan, K. H. Whitmire,
Chem. Mater. 2003, 15, 4407.
13] R. E. Bachman, K. H. Whitmire, J. H. Thurston, A. Gulea,
O. Stavila, V. Stavila, Inorg. Chim. Acta 2003, 346, 249.
14] K. H. Whitmire, J. C. Hutchison, A. Gardberg, C. Edwards,
Inorg. Chim. Acta 1999, 294, 153.
netic behavior of compounds 3, 4, and 5 resembles that of
related trinuclear compounds [19Ϫ24]. At room tempera-
3
Ϫ1
ture χ T is 1.49, 1.30 and 1.55, cm ·mol ·K for 3, 4 and
M
5
, respectively. These values are somewhat larger than those
II
expected for a magnetically uncoupled trinuclear Cu com-
3
Ϫ1
pound (1.23 cm ·mol ·K, given a g ϭ 2.1). Upon cooling
the samples, χ T increases and reaches a maximum at
M
[15] S. R. Breeze, S. Wang, J. E. Greedan, N. P. Raju, Inorg. Chem.
1996, 35, 6944.
[16] G. M. Sheldrick, SADABS, A program for Crystal Data
Absorption Correction, Univ. Göttingen, 1997.
[17] TEXSAN v. 5.0, Single Crystal Structure Analysis Software,
Molecular Structure Company, 1990.
[18] G.M. Sheldrick, SHELXTL, A program for the Solution and
Refinement of Crystal Structures, Univ. Göttingen, 2001.
3
Ϫ1
1
1
5
2.5 K of 2.15 cm ·mol ·K for 3, at 20 K of
.86 cm ·mol ·K for 4 and at 9 K of 2.32 cm ·molϪ1·K for
3
Ϫ1
3
. This behavior is characteristic of strong ferromagnetic
II
exchange coupling between the adjacent Cu atoms in trin-
II
uclear Cu species. Below these maxima the χ T product
M
Ϫ1
3
decreases rapidly down to 1.76 cm ·mol ·K (3),
.11 cm ·mol ·K (4) and 2.16 cm ·mol ·K (5),which is
3
Ϫ1
3
Ϫ1
1
[
19] S. Gehring, P. Fleischhauer, H. Paulus, W. Haase, Inorg. Chem.
993, 32, 54.
most likely due to ZFS within the quadruplet ground state.
1
[
20] P. Chaudhuri, M. Winter, B. P. C. Della Vedova, E. Bill, A.
Trautwein, S. Gehring, P. Fleischhauer, B. Nuber, J. Weiss,
Inorg. Chem. 1991, 30, 2148.
Conclusions
Heteroleptic complexes of copper(II) carboxylates with tri-
ethanolamine are readily obtained by direct reaction of
Cu(O CR) with TEAH or [{Cu(TEAH)} ] in stoichio-
metric amounts. By control of the reaction conditions,
mononuclear, dinuclear or trinuclear complexes may be ob-
tained.
[21] P. Fleischhauer, S. Gehring, C. Saal, W. Haase, Z. Tomkowicz,
C. Zanchini, D. Gatteschi, D. Davidov, A. L. Barra, J. Mag.
Mag. Mater 1996, 159, 166.
2
2
3
4
[22] S. Gehring, H. Astheimer, W. Haase, J. Chem. Soc., Faraday
Trans. 1987, 83, 347.
[23] S. Gehring, W. Haase, H. Paulus. Acta Crystallogr. 1991,
C47, 1814.
[
[
24] W. Haase, S. Gehring J. Chem. Soc., Dalton Trans 1985, 2609.
25] L. Gutierrez, G. Alzuet, J. A. Real, J. Cano, J. Borras, A.
Castineiras. Inorganic Chem. 2000, 39, 3608.
26] C. Y. Shu, J. M. Shi, L. D. Liu, C. J. Wu, M. R. Wang, P.
Cheng. Polish J. Chem. 2000, 74, 729.
Acknowledgments. The authors are grateful to the Robert A. Welch
Foundation (C-0976) for support of this work.
[
[
References
27] MAGMUN, http://www.ucs.mun.ca/ϳlthomp/magmun.html,
Memorial University of Newfoundland, St John’s, Newfound-
land,
[1] K. G. Caulton, L. G. Hubert-Pfalzgraf, Chem. Rev. 1990, 90,
969.
2876
 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 2867Ϫ2876