Dinuclear complexes and a one-dimensional chain involving difunctional ligands containing the acetylacetonate functionality

Article abstract of DOI:10.1007/s10870-007-9224-7

Systems containing two acetylacetonate (acac) ligands separated by an aromatic spacer (1-3) react with Cu(II) in the presence of bidentate nitrogen ligands to form crystalline, capped, dinuclear (zero-dimensional) structures (6-8). Noncovalent interaction

Full text of DOI:10.1007/s10870-007-9224-7

J Chem Crystallogr (2007) 37:629–639
DOI 10.1007/s10870-007-9224-7
ORIGINAL PAPER
Dinuclear Complexes and a One-dimensional Chain Involving
Difunctional Ligands Containing the Acetylacetonate
Functionality
Joseph B. Lambert Æ Zhongqiang Liu
Received: 30 November 2006 / Accepted: 18 June 2007 / Published online: 24 July 2007
ꢀ Springer Science+Business Media, LLC 2007
Abstract Systems containing two acetylacetonate (acac)
ligands separated by an aromatic spacer (1–3) react with
Cu(II) in the presence of bidentate nitrogen ligands to form
crystalline, capped, dinuclear (zero-dimensional) structures
(6–8). Noncovalent interactions between stacked aromatic
ligands on adjacent molecules in 7 and 8 produce one-
dimensional chains in the crystal. Molecules 1–3 fail to
produce crystalline, one-dimensional chains with covalent
metal-organic bonding on reaction with Cu(II) alone.
System 5, which contains a single acac ligand and a pyridyl
nitrogen, forms a crystalline, one-dimensional chain on
reaction with Cu(II) in the presence of a bidentate nitrogen
ligand (9). The covalent bonding ribbon of the chain passes
from acac to Cu(II) to the pyridyl nitrogen and back to
acac. The phenanthroline ligands in 9 provide noncovalent
bonding with adjacent chains in the crystal to produce a
two-dimensional layered structure.
bonds. The low energy of the hydrogen bond, however,
leads to two negative factors not shared by metal-organic
bonds. First, during the process of recrystallization,
hydrogen bonds can form, break, and reform, with the
result that multiple, interpenetrating, multidimensional
structures can grow around and through each other and
fill crystalline voids. The resulting materials are non-
porous. In one example from our own work, the tetra-
carboxylic acid Si[4-(CO2H)C6H4]4 crystallizes as six
separate, interpenetrating hydrogen-bonded networks [6].
Second, even when voids are left during crystallization,
they usually are filled with solvent or other guests. At-
tempts at thermal removal of the guests often result in
collapse of a crystal based on hydrogen bonding con-
nections, with concomitant loss of porosity. Architectures
based on hydrogen bonds thus are not ideally suited for
forming robust, porous solids, which may be suitable for
catalysis and separations.
Keywords Acetylacetonate ligands Á Aromatic stacking Á
Noncovalent bonding Á One-dimensional chains
The intermediate strength of many metal-organic bonds
has made them the frequent basis for the formation of
porous materials such as metal-organic frameworks [1–3].
The carboxylic acid functionality again is the most com-
mon organic connector, but now in its ionized carboxylate
form, bonded to a metal. When translated into two and
three dimensions, such motifs can create complex struc-
tures containing significant voids. Such systems often are
robust to heating, for example, during the process of sol-
vent removal. Thus metal-organic frameworks have proved
to be superior to hydrogen-bonded networks in creating
solids with large voids.
Introduction
Metal-organic bonds [1–3], like hydrogen bonds [4, 5],
are used commonly as the basis for the formation of
organized solids that are held together by noncovalent
Electronic supplementary material The online version of this
article (doi:10.1007/s10870-007-9224-7) contains supplementary
material, which is available to authorized users.
The vast majority of metal-organic frameworks has
contained the carboxylate building block, and the
remaining examples usually contained pyridine function-
alities. In 1984, Maverick and co-workers [7–9] reported
that organic-connected acetylacetonate (acac) ligands
J. B. Lambert (&) Á Z. Liu
Department of Chemistry, Northwestern University, 2145
Sheridan Road, Evanston, IL 60208-3113, USA
e-mail: jlambert@northwestern.edu
123

630
J Chem Crystallogr (2007) 37:629–639
form dinuclear complexes with copper in macrocyclic
structures reminiscent of the pyridine-based structures of
Fujita, Stang, and others [10]. By placing two acac groups
onto a single molecule with a bent geometry, Maverick
et al. prepared dimers linked by copper ions. These dimers
proved to be soluble in organic solvents and capable of
serving as molecular hosts. A pair of acac ligands (A) bear
a formal similarity to a pair of carboxylates (B) in their
copper complexes. The group R serves as a connector to a
second acac ligand, so that either cyclic or infinite arrays
could be formed. Maverick’s bent molecules [7–9] led to
cyclic (finite) arrays, which might be termed the zeroth
order structure of a topological family.
Experimental
Chemicals
4,4¢-Diformylbiphenyl
[17],
2,2,2-trimethoxy-4,5-di-
methyl-1,3-dioxaphospholene [18, 19], and 4-(3¢-acetyl-
acetonato)pyridine (5) [15] were synthesized according to
the literature and were characterized by GC-MS and NMR
spectra, which matched the reported data. Other cited
materials were available commercially.
Synthesis of Monomers
1,4-Di-(3¢-acetylacetonato)benzene (1) [20, 21]
O
O
O
O
A 250 mL flask containing a magnetic stirrer and tereph-
thalaldehyde (13.4 g, 0.1 mol) was placed in an ice bath,
subjected to vacuum, and refilled with nitrogen for three
cycles. Anhydrous dichloromethane (100 mL) was trans-
ferred into the flask by a cannula. Freshly prepared
2,2,2-trimethoxy-4,5-dimethyl-1,3-dioxaphospholene
(50 g, 0.24 mol) was added dropwise. The ice bath was
removed, and the mixture was stirred overnight. The vol-
atiles were removed by vacuum, and anhydrous CH3OH
(150 mL) was admitted through a cannula. The mixture
was heated to reflux for 24 h and cooled. The volatiles
were removed under vacuum, and the solid was recrystal-
lized in benzene to give a white solid: 18.5 g (67%); GC-
MS, m/z 274 (M+); NMR dH (400 MHz; CDCl3; Me4Si)
16.68 (2 H, s), 7.20 (4 H, s), 1.91 (12 H, s); dC (100 MHz;
CDCl3; Me4Si) 190.7, 136.3, 131.6, 114.8, 24.4.
Cu
R
R
A
O
O
O
O
Cu
R
R
B
The structural key is the presence of two acac groups in
the same molecule. The most studied example of this motif
also has been the simplest, tetraacetylethane (C, shown in
its dihydroxy tautomer), in which the two acac groups are
attached to each other directly at their 3 positions [11–13].
In one case [11] noncrystalline materials were obtained
with Cu(II), in a second example [12] a molecular square
was obtained with Co(II), and in a third [13] a hydrogen-
bonded chain structure was obtained with Cu(II) and di-
pyridyl co-reactants. Domasevitch [14], Maverick [15, 16],
and their co-workers have used acac linked to pyridine to
produce two-dimensional metal-organic polymers. In the
present study, we used linear molecules containing acac
and a second ligating functionality to explore whether acac
can participate in dinuclear and higher order (infinite) ar-
rays.
1,4-Di-[methylene-3¢-(acetylacetonato)]benzene (2)
[22, 23]
To a stirred mixture of chlorotrimethylsilane (25.6 mL,
0.2 mol), NaI (30 g, 0.2 mol), 2,4-pentanedione (4 g,
0.04 mol), and acetonitrile (200 mL) in a 500 mL flask at 0
ꢁC was added dropwise an acetonitrile solution of tereph-
thalaldehyde (2.68 g, 0.02 mol). The mixture was stirred
for 10 h at room temperature and then for 24 h at 68 ꢁC.
After addition of water (200 mL), the mixture was ex-
tracted with ether. The combined ether extracts were wa-
shed with aqueous Na2S2O3 solution to remove the
liberated iodine, washed with brine and fresh water, and
dried (MgSO4). Solvents were removed by rotary evapo-
ration at room temperature. The white solid was recrys-
tallized from tert-butanol to give white plate-like crystals:
5.0 g (83%). Spectra were in agreement with the literature
[23].
O
OH
O
HO
C
123

J Chem Crystallogr (2007) 37:629–639
631
4,4¢-Di-(3¢-acetylacetonato)biphenyl (3)
placed into a water bath (32 ꢁC). The mixture was stirred
for 4 h, and the volume of the solution was reduced under
vacuum. The resulting solid was extracted by ether. The
product was purified by flash chromatography (hexane/
ether) to give a white solid: 140 mg (75%); mp 197.5 ꢁC;
GC-MS, m/z 220 (M+); NMR, dH (400 MHz; acetone-d6;
Me4Si) 16.90 (1 H, s), 8.10 (2 H, d), 7.46 (2 H, d), 1.90 (6
H, s); dC (100 MHz; acetone-d6; Me4Si) 191.6, 167.5,
142.8, 132.4, 130.9, 129.6, 115.4, 24.3. Anal. Calcd.
(found) for C12H12O4: C 64.45 (65.39); H 5.49 (5.47).
A 250 mL flask containing a magnetic stirrer and 4,4¢-
diformylbiphenyl (5 g, 24 mmol) was placed in an ice
bath, subjected to vacuum, and refilled with nitrogen for
three cycles. Anhydrous dichloromethane (50 mL) was
transferred into the flask by a cannula. Freshly prepared
2,2,2-trimethoxy-4,5-dimethyl-1,3-dioxaphospholene
(15 g, 71 mmol) was added dropwise. The ice bath was
removed, and the mixture was stirred overnight. The vol-
atiles were removed under vacuum, and anhydrous CH3OH
(50 mL) was admitted through a cannula. The mixture was
heated to reflux for 24 h and cooled. The volatiles were
removed by rotary evaporation, and the residue was dried
at 80 ꢁC (0.5 torr) for 24 h. A white solid was collected,
and the remaining residue was discarded. The solid was
recrystallized from benzene to give white crystals: 1.5 g
(18%); GC-MS, m/z 350 (M+); NMR, dH (400 MHz;
CDCl3; Me4Si) 16.72 (1 H, s), 7.67 (4 H, m), 7.28 (4 H, m),
1.95 (12 H, s); dC (100 MHz; CDCl3; Me4Si) 191.1, 139.7,
136.4, 131.8, 127.6, 115.0, 24.6. Anal. Calcd. (found) for
C22H22O4: C, 75.41 (75.49); H, 6.33 (6.41); N, 0.00 (0.01).
Synthesis of Polymers
Attempted Chelate Polymer Formation
of 1,4-di-(3¢-acetylacetonato)benzene (1),
1,4-di-[methylene-3¢-(acetylacetonato)]benzene (2),
and 4-(3¢-acetylacetone)benzoic acid (4)
To a 250 mL flask containing a magnetic stirrer was added
1,4-di-[methylene-3¢-(acetylacetonato)]benzene (2) (0.6 g,
2 mmol) in dichloromethane (15 mL). Tetraamminecop-
per(II) sulfate (0.46 g, 2 mmol) in water (15 mL) was
added dropwise [25, 26]. After the solution had been stirred
for 30 min, a gray, noncrystalline precipitate was collected
by filtration and washed with water. Chelate polymeriza-
tions of 1,4-di-(3¢-acetylacetonato)benzene (1) and of 4-
(3¢-acetylacetonato)benzoic acid (4) with tetraammine-
copper(II) sulfate were carried out in a similar fashion [27].
The products were amorphous materials that did not dis-
solve in water or in common organic solvents. Single
crystals of these materials failed to grow by slow diffusion
of the reactants in two solutions.
Methyl 4-(3¢-acetylacetonato)benzoate
A 250 mL flask containing a magnetic stirrer and methyl
4-formylbenzoate (11 g, 67 mmol) was placed in an ice
bath, subjected to vacuum, and refilled with nitrogen for
three cycles. Anhydrous dichloromethane (150 mL) was
transferred into the flask by a cannula. Freshly prepared
2,2,2-trimethoxy-4,5-dimethyl-1,3-dioxaphospholene
(28 g, 133 mmol) was added dropwise. After 2 h, the ice
bath was removed, and the solution was stirred overnight.
The volatiles were removed by vacuum, and anhydrous
CH3OH (150 mL) was admitted through a cannula. The
mixture was heated to reflux for 3 h. The volatiles were
removed by rotary evaporation, and the residue was dried
at 80 ꢁC (0.5 torr) for 24 h. A white solid was collected,
and the remaining residue was discarded. The solid was
recrystallized from benzene to give white crystals: 8.1 g
(55%); GC-MS, m/z 234 (M+); NMR, dH (400 MHz;
CDCl3; Me4Si) 16.66 (1 H, s), 8.01 (2 H, d), 7.23 (2 H,
d), 3.88 (3 H, s), 1.85 (6 H, s); dC (100 MHz; CDCl3;
Me4Si) 190.8, 166.9, 142.1, 131.5, 130.3, 129.6, 114.5,
52.4, 24.3.
Dinuclear Copper Complex of 1,4-di-(3¢-
acetylacetonato)benzene with Tetramethylethylenediamine
(6)
To a 50 mL flask containing a magnetic stirrer and acetone
(10 mL) was added dropwise first copper perchlorate
hexahydrate (185 mg, 0.5 mmol) and then tetramethyle-
thylenediamine (58 mg, 0.5 mmol) in acetone (10 mL).
After addition was complete, the mixture was stirred for
another 5 min, and 1,4-di-(3¢-acetylacetonato)benzene (1,
69 mg, 0.25 mmol) in acetone (5 mL) was added slowly
and dropwise. The mixture was stirred for another 10 min,
and NaOH (20 mg) in CH3OH was added. A blue precipi-
tate appeared. The mixture was subjected to ultrasonic
sound for several hours until the solution became clear blue.
The solution was reduced to 3 mL under vacuum. Ether was
added, and a dark blue precipitate was collected and dried
by vacuum: 154 mg (74%). A sample of the product
(30 mg) was dissolved in the minimum amount of acetone
4-(3¢-Acetylacetonato)benzoic acid (4) [24]
A 100 mL flask containing a magnetic stirrer, 4-(3¢-acet-
ylacetonato)benzoate (0.2 g, 0.85 mmol) in acetone
(5 mL), pig liver esterase (60 mg, 1440 units), and a buffer
solution (50 mL, 0.05 M, pH = 6.5) of KH2PO4/KOH was
123

632
J Chem Crystallogr (2007) 37:629–639
and was transferred to a small vial, which was placed in a
large capped vial with ether/dimethylformamide. After
several days, dark purple crystals were collected. Anal.
Calcd. (found) for (C16H16O4)Cu2(C6H16N2)2(ClO4)2: C,
40.48 (40.63); H, 5.82 (5.71); N, 6.74 (6.50).
24 h to remove acetone and DMSO prior to elemental
analysis. Anal. Calcd. (found) for (C16H16O4)Cu2(C10H8
N2)2(ClO4)2: C, 47.48 (47.34); H, 3.54 (3.50); N, 6.15
(5.97).
Dinuclear Copper Complex of 4,4¢-di-(3¢-
acetylacetonato)biphenyl with 2,2¢-bipyridine (8)
Dinuclear Copper Complex of 1,4-di-(3¢-
acetylacetonato)benzene with 2,2¢-bipyridine (7)
To a 50 mL flask containing a magnetic stirrer and acetone
(10 mL) was added dropwise first copper perchlorate
hexahydrate (185 mg, 0.5 mmol) and then an acetone
solution (10 mL) of 2,2¢-bipyridine (78 mg, 0.5 mmol).
After addition was complete, the mixture was stirred for
another 5 min and 4,4¢-di-(3¢-acetylacetonato)biphenyl
(87 mg, 0.25 mmol) in acetone (5 mL) was added slowly
and dropwise. The mixture was stirred for another 10 min,
and NaOH (20 mg) in CH3OH was added. The precipitate
was collected and dried under vacuum: 204 mg (83%). A
portion of the product (30 mg) was dissolved in the mini-
mum amount of dimethylformamide (DMF) and was
transferred to a small vial. The vial was put in a large
capped vial with ether/DMF. Four days later, dark blue
single crystals were collected. Anal. Calcd. (found) for
(C22H20O4)Cu2(C10H8N2)2(C3H7NO)1.5(ClO4)2: C, 50.94
(50.81); H, 4.27 (4.39); N, 7.03 (7.03).
To a 50 mL flask containing a magnetic stirrer and ace-
tone (10 mL) was added dropwise first copper perchlorate
hexahydrate (185 mg, 0.5 mmol) and then an acetone
solution (10 mL) of 2,2¢-bipyridine (78 mg, 0.5 mmol).
After addition was complete, the mixture was stirred for
another 5 min, and 1,4-di-(3¢-acetylacetonato)benzene (1,
69 mg, 0.25 mmol) in acetone (5 mL) was added slowly
and dropwise. The mixture was stirred for another 10 min,
and NaOH (20 mg) in CH3OH was added. The precipitate
was collected and dried by vacuum: 195 mg (86%). A
portion of the product (30 mg) was dissolved in the
minimum amount of dimethyl sulfoxide (DMSO) and was
transferred to a small vial. The vial was placed in a large
capped vial with acetone/DMSO. One week later, dark
blue single crystals were collected for X-ray crys-
tallography. The crystals were dried under vacuum for
Table 1 Crystal data and
Identification code
92q
structure refinement for 6
Empirical formula
Formula weight
C56 H96 Cl4 Cu4 N8 O24
1661.37
Temperature
153(2) K
˚
0.71073 A
Wavelength
Crystal system, space group
Unit cell dimensions
Monoclinic, C2/c
˚
˚
a = 27.8046(17) A, b = 9.7529(6) A,
˚
b = 90.3700(10)ꢁ, c = 13.2329(8) A
3
˚
Volume
3588.4(4) A
Z, Calculated density
Absorption coefficient
F(000)
2, 1.538 mg/m3
1.397 mm–1
1728
Crystal size
0.452 · 0.442 · 0.088 mm
1.46 to 29.17ꢁ
–37 £ h £ 37, –12 £ k £ 13, –18 £ l £ 18
21531/4473 [R(int) = 0.0613]
92.1%
Theta range for data collection
Limiting indices
Reflections collected/unique
Completeness to theta = 29.17
Absorption correction
Transmission factors
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2sigma(I)]
R indices (all data)
Integration
0.8792 (maximum) and 0.5466 (minimum)
Full-matrix least-squares on F2
4473/0/223
1.033
R1 = 0.0355, wR2 = 0.0909
R1 = 0.0446, wR2 = 0.0939
–3
˚
Largest diff. peak and hole
0.454 and –0.487 e-/A
123

J Chem Crystallogr (2007) 37:629–639
633
Table 2 Crystal data and
structure refinement for 7
Identification code
s51q1m
Empirical formula
Formula weight
C23 H28 Cl Cu N2 O8 S
591.52
Temperature
293(2) K
˚
0.71073 A
Wavelength
Crystal system, space group
Unit cell dimensions
Triclinic, P-1
˚
˚
a = 11.0425(9) A, a = 104.8720(10)ꢁ, b = 11.3196(10) A,
˚
b = 101.6620(10)ꢁ, c = 12.0961(10) A, c = 108.1070(10)ꢁ
3
˚
1321.94(19) A
Volume
Z, Calculated density
Absorption coefficient
F(000)
2, 1.486 Mg/m3
1.055 mm–1
612
Crystal size
0.322 · 0.214 · 0.082 mm
1.83 to 29.28ꢁ
–14 £ h £ 14, –15 £ k £ 14, –16 £ l £ 15
11965/6254 [R(int) = 0.0657]
86.6%
Theta range for data collection
Limiting indices
Reflections collected/unique
Completeness to theta = 29.28
Absorption correction
Transmission factors
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2sigma(I)]
R indices (all data)
Integration
0.9221 (maximum) and 0.7400 (minimum)
Full-matrix least-squares on F2
6254/0/331
1.094
R1 = 0.0762, wR2 = 0.2119
R1 = 0.1124, wR2 = 0.2733
–3
˚
Largest diff. peak and hole
1.331 and –1.360 e-/A
One-dimensional Polymer
of 4-(3¢-acetylacetonato)pyridine (9)
with 1,10-phenanthroline and Cu(II)
para) of a connector (Scheme 1). Such a juxtaposition is
intended to discourage finite cyclic arrays in favor of
infinite linear arrays. In contrast to tetraacetylethane (C),
the acac groups here are separated by a spacer: 1,4-
phenylene in 1, 1,4-phenylenebismethylene in 2, and
4,4¢-biphenylene in 3. Molecule 4 contains acac and the
carboxylic acid functionality separated by 1,4-phenylene,
and 5 contains acac para to a pyridine nitrogen [14–16].
Reaction with Cu(II) salts failed to yield crystalline
polymer. Only amorphous materials formed, as previ-
ously observed for 1 [20]. The simple phenylene com-
pound 1, however, reacted with Cu(II) in the presence of
bidentate nitrogen bases (either N,N,N¢,N¢-tetramethyle-
thane-1,2-diamine or 2,2¢-bipyridine) to form dinuclear
complexes, 6 (Fig. 1) and 7 (Fig. 2). No useful products
were isolated from the substrate 2 in the presence of a
chelate. The biphenylene substrate 3 also produced a
dinuclear complex, 8 (Fig. 3), on reaction with 2,2¢-bi-
pyridine. These finite complexes are considered to be
zero dimensional. No materials were isolated from the
reaction of the mixed carboxylate/acac substrate 4 in the
presence of a chelate. The mixed pyridine/acac substrate
5 formed an infinite one-dimensional zigzag chain with
the unit structure 9 (Fig. 4) in the presence of 1,10-
phenanthroline.
Copper perchlorate hexahydrate (91 mg, 0.246 mmol) and
1,10-phenanthroline (45 mg, 0.25 mmol) were dissolved in
a combinatory solution of CH3OH (15 mL) and DMF
(5 mL). The clear blue-green solution was subjected to
ultrasonic sound for 10 min. 4-(3¢-Acetylacetonato)pyri-
dine (5, 44 mg, 0.25 mmol) was dissolved in CH3OH
(20 mL). In five glass tubes, the CH3OH solution (4 mL) of
the acetylacetone derivative 5 was carefully layered upon
the combinatory solution. After one week, green crystals
appeared at the interface and fell to the bottom.
Crystal Structures
Crystal data are presented in Table 1 for 6, Table 2 for 7,
Table 3 for 8, and Table 4 for 9. Experimental details are
given in the Supplementary Materials.
Results and Discussion
We prepared five molecules that contain one or two acac
functionalities that are positioned at opposite ends (1,4 or
123

634
J Chem Crystallogr (2007) 37:629–639
Table 3 Crystal data and
structure refinement for 8
Identification code
Empirical formula
Formula weight
s81q1m
C24 H25 Cl Cu N3 O7
566.46
Temperature
153(2) K
˚
0.71073 A
Wavelength
Crystal system, space group
Unit cell dimensions
Triclinic, P-1
˚
˚
a = 10.832(3) A, a = 88.258(5)ꢁ, b = 10.876(3) A,
˚
b = 83.332(5)ꢁ, c = 11.551(4) A, c = 69.298(4)ꢁ
3
˚
Volume
1264.2(7) A
Z, Calculated density
Absorption coefficient
F(000)
2, 1.488 Mg/m3
1.018 mm–1
584
Crystal size
0.410 · 0.376 · 0.030 mm
1.78 to 28.58ꢁ
–14 £ h £ 13, –14 £ k £ 14, –14 £ l £ 15
11286/5738 [R(int) = 0.0253]
88.7%
Theta range for data collection
Limiting indices
Reflections collected/unique
Completeness to theta = 28.58
Absorption correction
Transmission factors
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2sigma(I)]
R indices (all data)
Integration
0.9696 (maximum) and 0.6891 (minimum)
Full-matrix least-squares on F2
5738/0/336
1.041
R1 = 0.0446, wR2 = 0.1173
R1 = 0.0584, wR2 = 0.1282
–3
˚
Largest diff. peak and hole
0.983 and –0.673 e-/A
Table 4 Crystal data and
structure refinement for 9
Identification code
Empirical formula
s20r1m
C22 H18 Cl Cu N3 O6
519.38
Formula weight
Temperature
153(2) K
˚
0.71073 A
Wavelength
Crystal system, space group
Unit cell dimensions
Volume
Orthorhombic, Cmca
˚
˚
a = 11.3015(8) A, b = 25.3610(18) A, c = 15.2995(11) A
˚
3
˚
4385.1(5) A
Z, Calculated density
Absorption coefficient
F(000)
8, 1.573 Mg/m3
1.163 mm–1
2120
Crystal size
0.160 · 0.142 · 0.082 mm
1.61 to 28.70ꢁ
–14 £ h £ 14, –32 £ k £ 34, –20 £ l £ 19
19818/2858 [R(int) = 0.1070]
96.1%
Theta range for data collection
Limiting indices
Reflections collected/unique
Completeness to theta = 28.70
Absorption correction
Transmission factors
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2sigma(I)]
R indices (all data)
Largest diff. peak and hole
Integration
0.9194 (maximum) and 0.8152 (minimum)
Full-matrix least-squares on F2
2858/0/218
1.033
R1 = 0.0457, wR2 = 0.1194
R1 = 0.0758, wR2 = 0.1353
–3
˚
0.587 and –0.929 e-/A
123

J Chem Crystallogr (2007) 37:629–639
635
Scheme 1
OH
HO
O
O
O
O
OH
HO
1
2
HO
O
O
OH
3
O
O
HO
O
N
HO
OH
5
4
O
ClO
2
4
O
Cl
O
N
N
O
N
O
O
O
Cu
Cu
O
N
N
O
O
O
N
Cu
Cu
N
N
N
Cl
O
O
O
O
6
O
N
lCO
2
4
S
9
N
N
N
O
O
O
Cu
Cu
O
The acetylacetonate functionality, in contrast to car-
O
O
N
boxylate, exhibited little ability in this study to form infi-
nite chains as isolable, crystalline substances. Crystal
formation of infinite chains or networks is a delicate
function of the strengths of the bonds forming between the
organic and inorganic components. Reversibility is a crit-
ical characteristic of the crystallization process, in order
that errors in the crystal may be repaired. When bonds that
connect portions of the solid are quite strong, incorrect
structures (those that do not match the overall crystal lat-
tice) cannot be repaired. Such errors, when propagated,
lead to amorphous materials. One measure of bond strength
in the present context is the stability constant between the
metal ion and the organic ligand. For extremely large
S
7
Me2N
ClO
2
4
C
O
H
N
N
N
O
O
O
Cu
Cu
O
N
O
H
C
NMe2
8
123

636
J Chem Crystallogr (2007) 37:629–639
Fig. 1 The crystal structure of 6 (50% probability level thermal
ellipsoids)
constants, the dissociation rate is much slower than the
association rate. As a result, incorrectly formed substruc-
tures that cannot be repaired promote noncrystallinity. The
binding constant of copper to carboxylate is about six or-
ders of magnitude smaller than that of copper to acetyl-
acetonate [28]. This difference provides an explanation for
the inability of acac to form infinite chains or networks in
this study, whereas similar structures are common for the
carboxylate function [29].
Fig. 2 The crystal structure of 7 (50% probability level thermal
ellipsoids)
Such limitations do not apply to single crystals of small
complexes, such as the dinuclear complexes 6–8 that were
prepared in this work. The bidentate nitrogen ligands were
necessary to cap the copper centers. Rather than reacting
with acac in another molecule to form oligomers and
eventually amorphous materials containing infinite chains,
the copper centers reacted with the nitrogen base and
crystallized as small molecules. All these materials have
solvent or anion as a fifth coordination on copper. This
result suggested to us that higher coordination could offer a
mechanism for the successful formation of crystalline,
infinite chains. Molecule 5 has been studied in this context
[14–16]. The pyridine ligand might fulfill the same role as
solvent/anion in 6–8. If so, infinite chains could be formed
that included acac as a part of the chain. Domasevitch and
co-workers [14] obtained a mixed metal Cu(Co)/Be two-
dimensional array with this ligand. Maverick and co-
workers [15, 16] obtained Cu and Cu/Co two-dimensional
arrays with this ligand. A novel one-dimensional chain was
realized in the present study with 9, in which the chain
leads from acac to copper (ligated to the bidentate phe-
nanthroline) to the pyridyl nitrogen and back to acac at the
other end of the pyridine ring.
Complexes 7–9, which contain the bipyridine or phe-
nanthroline ligands, exhibit interesting noncovalent inter-
actions in the solid. By means of p–p stacking or distant p-
Cu coordination, these complexes exist as higher order
arrays. Figures 5 and 6 show the packing of molecular
pairs in the dinuclear complexes 7 and 8. In both cases, the
bipyridine group is arranged parallel to the bipyridine
group of another molecule and is offset slightly in order to
Fig. 3 The crystal structure of 8 (50% probability level thermal
ellipsoids)
fill the role of a distant sixth ligand for nearly octahedral
copper. The square formed by the two acac oxygens and
the two bipyridyl nitrogens around copper in one molecule
is parallel to the analogous square in the adjacent molecule.
123

J Chem Crystallogr (2007) 37:629–639
637
Fig. 4 The crystal structure of 9 (50% probability level thermal
ellipsoids). Labels for the disordered phenathroline ring are given in
an expansion below the structure
In both cases, noncovalent bonding between the squares
and the bipyridine ligands on one molecule with the same
entities on the parallel molecule generate an infinite one-
dimensional chain in the solid. Such a material based on
noncovalent bonded does not qualify as a metal-organic
framework.
The phenanthroline ligands in the one-dimensional
chain of 9 also engage in noncovalent binding with an
adjacent chain, as illustrated in Fig. 7, which contains
portions of two zigzag chains. There is some disorder in the
phenanthroline pieces, but it is clear that the structure in the
crystal is controlled by p–p stacking between two aromatic
rings. The zigzag chain is arranged in such a way that
alternate phenanthrolines (let us call them the even ligands)
face in one direction to bind noncovalently with alternate
phenanthrolines on the adjacent chain. Then the odd li-
gands on that chain bind with an adjacent chain in the
opposite direction. The result is a planar, two-dimensional
array held together by noncovalent interactions.
Fig. 5 Crystal packing of 7, showing p–p stacking between adjacent
molecules to create a noncovalent, one-dimensional chain. Copper
(cyan), nitrogen (blue), carbon (gray), oxygen (red), and sulfur
(yellow)
of so-called Janus molecules) fail to form crystalline metal-
organic frameworks with Cu(II) salts in this study. We
attribute this result to the relatively strong oxygen-copper
bonds in acac complexes, in comparison with carboxylate
systems, which prohibit the free formation, breaking, and
reformation of bonds during crystallization. In the presence
of bidentate nitrogen ligands, these systems form dinuclear,
capped systems (6–8) (organic/copper ratio of 1/2), in
which copper has distorted pentacoordination to two
Summary
Aromatic systems (1–3) containing two acetylacetonate
(acac) substituents facing in opposite directions (examples
123

638
J Chem Crystallogr (2007) 37:629–639
Fig. 6 Crystal packing of 8 showing p–p stacking between adjacent
molecules to create a noncovalent, one-dimensional chain. Copper
(cyan), nitrogen (blue), carbon (gray), and oxygen (red)
Fig. 7 Crystal packing of 9, showing p–p stacking between adjacent
molecules to create a noncovalent layer. Copper (cyan), nitrogen
(blue), carbon (gray), and oxygen (red)
oxygens of acac, two nitrogens of the amine ligand, and
one oxygen of solvent or anion. These may be considered
zero-dimensional systems. Complexes 7 and 8 include bi-
pyridine as the bidentate nitrogen ligand. In the crystal, an
aromatic ring of one molecule stacks with an aromatic ring
of another, providing a noncovalently bound sixth ligand
for copper. Since each molecule contains two bipyridines,
at opposite ends, aromatic stacking results in an infinite
one-dimensional chain in the crystal, held together by these
noncovalent bonding arrangements (Fig. 5 and 6).
Because copper complexes 7 and 8 contain fifth coor-
dination from solvent, we considered whether a modified
acac complex, in which the second acac of 1–3 is replaced
by a more loosely coordinating pyridine entity, could form
covalent one-dimensional chains. The dual functionalities
of molecule 5 thus are acac and the pyridyl nitrogen.
123

J Chem Crystallogr (2007) 37:629–639
639
7. Maverick AW, Klavetter FE (1984) Inorg Chem 23:4129
8. Maverick AW, Buckingham SC, Bradbury JR, Yao Q, Stanley
GG (1986) J Am Chem Soc 108:7430
9. Maverick AW, Ivie ML, Waggenspack JW, Fronzek FR (1990)
Inorg Chem 29:2403
10. Stang PJ, Olenyuk B (1997) Acc Chem Res 30:502; Fujita M
(1999) Acc Chem Res 32:53
11. Shi JM, Xu JQ, Wang RZ, Yang GY, Sun HR, Wang TG, Cheng
P, Liao DZ (1998) Polish J Chem 72:1273
Indeed, reaction of 5 with Cu(II) salts produced a covalent,
infinite, one-dimensional chain in which the polymerizing
ribbon included acac (9). This material contained phenan-
throline to provide two ligands on copper. Noncovalent
bonding of phenanthroline rings on adjacent chains creates
a two-dimensional layered structure (Fig. 7). Interpenetra-
tion is not present in any of these structures.
12. Zhang Y, Wang S, Enright GD, Breeze SR (1998) J Am Chem
Soc 120:9398
13. Zhang Y, Breeze SR, Wang S, Greedan JE, Raju NP, Li L (1999)
Can J Chem 77:1424
Supplementary Materials
14. Vreshch VD, Chernega AN, Howard JAK, Sieler J, Domasevitch
KV (2003) J Chem Soc Dalton Trans 1707–1711
15. Chen B, Fronczek FR, Maverick AW (2003) Chem Commun
2166–2167
16. Chen B, Fronczek FR, Maverick AW (2004) Inorg Chem 43:8209
17. Demir AS, Reis O, Emrullahoglu M (2003) J Org Chem 68:10130
18. Ramirez F, Ramanathan N, Desai NB (1962) J Am Chem Soc
84:1317
19. Castelijns MMCF, Schipper P, Van Aken D, Buck HM (1981) J
Org Chem 46:47
20. Ramirez F, Bhatia SB, Patwardhan AV, Smith CP (1967) J Org
Chem 32:3547–3553
Crystallographic details for compounds 6–9. The .cif files
have been deposited as CCDC-619531, CCDC-619532,
CCDC-619533, and CCDC-619534. These data can be
obtained free of charge via www.ccdc.cam.ac.uk/data_re-
quest/cif, or by emailing data_request@ccdc.cam.ac.uk, or
by contacting The Cambridge Crystallographic Data Cen-
tre, 12 Union Road, Cambridge CB2 1EZ, UK; fax:
+44(0)1223-336033.
Acknowledgments We acknowledge support of the National Sci-
ence Foundation (CHE-0349412).
21. Tsuboyama K, Yanagita M (1967) Scientific Papers of the
Institute of Physical and Chemical Research (Japan) 61:20
22. Martin RF, Fernelius WC, Shamma M (1959) J Am Chem Soc
81:130
References
23. Sakai T, Miyata K, Tsuboi S, Utaka M (1989) Bull Chem Soc Jpn
62:4072
24. Jiang Y, Wu N, Wu H, He M (2005) Synlet 2731–2734
25. Liu Y, Liu C, Cui L, Fan Z, Xie P, Zhang R (2000) Liq Crist 27:5
26. Knobloch FW, Rauscher WH (1959) J Polym Sci 38:261
27. Charles RG (1960) J Phys Chem 64:1747
28. Sircar SC, Aditya S, Prasad B (1952) J Indian Chem Soc 30:633;
Bryant BE (1954) J Phys Chem 58:573
1. James SL (2003) Chem Soc Rev 32:276
2. Ockwig NW, Delgado-Friedrichs O, O’Keefe M, Yaghi OM
(2005) Acc Chem Res 38:176
3. Hupp JT, Poeppelmeier KR (2005) Science 309:2040
4. Lawrence DS, Jiang R, Levett M (1995) Chem Rev 95:2229
5. Duchamp DG, Marsh RE (1969) Acta Crystallogr Sect B: Struct
Sci 25:5
29. Martell AE (1964) Stability constants of metal-ion complexes.
Section II. Organic ligands. The Chemical Society, London
6. Lambert JB, Zhao Y, Stern (1997) J Phys Org Chem 10:229
123