A New Binucleating Ligand Based on Anthracene and Its Cofacial Dirhodium(I) and Diiridium(I) Complexes

Article abstract of DOI:10.1021/ic9706358

A new bis(β-keto enamine) ligand (2a, ABIH₂) containing a 1,8-anthracenediyl bridging group has been synthesized by a four-step procedure that relies on the Pd-catalyzed cross-coupling between (3,5-dimethylisoxazol-4-yl)-trialkyltin and 1,8-dibromoanthraquinone or -anthracene. The molecular structures of the 1,8-bis(3,5-dimethylisoxazol-4-yl)anthraquinone (8) and -anthracene (10) intermediates were determined by X-ray analysis. Crystal data for 8: orthorhombic, space group Pbca; a = 14.351 (2), b = 11.932 (1), c = 23.278 (2) A?; V = 3986 (1) A?³; Z = 8; R = 0.057 for 2615 reflections. Crystal data for 10: orthorhombic, space group P2₁2₁2₁; a = 7.104 (1), b = 12.805 (1), c = 22.280 (2) A?; V = 2026.7 (6) A?³; Z = 4; R = 0.066 for 3423 reflections. The rigid ABIH₂ ligand, whose chelating moieties are constrained to be cofacial, allows the preparation of a new family of cofacial bimetallic complexes (ABI)[ML₂]₂ with controllable environments around the metal centers. Two novel cofacial binuclear complexes 4 and 5, with ML₂ = dicarbonylrhodium(I) and (η⁴-1,5-cyclooctadiene)iridium(I), have been synthesized by reaction of ABIH₂ with [(μ-Cl)Rh(CO)₂]₂ and [(μ-Cl)Ir(COD)]₂, respectively. NMR data indicate the formation of meso and racemic atropisomers for 2a, 4, and 5.

Full text of DOI:10.1021/ic9706358

5826
Inorg. Chem. 1997, 36, 5826-5831
A New Binucleating Ligand Based on Anthracene and Its Cofacial Dirhodium(I) and
Diiridium(I) Complexes
Mar´ıa del Rosario Benites, Frank R. Fronczek, Robert P. Hammer, and
Andrew W. Maverick*
Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803
ReceiVed May 23, 1997X
A new bis(â-keto enamine) ligand (2a, ABIH2) containing a 1,8-anthracenediyl bridging group has been synthesized
by a four-step procedure that relies on the Pd-catalyzed cross-coupling between (3,5-dimethylisoxazol-4-yl)-
trialkyltin and 1,8-dibromoanthraquinone or -anthracene. The molecular structures of the 1,8-bis(3,5-dimethyl-
isoxazol-4-yl)anthraquinone (8) and -anthracene (10) intermediates were determined by X-ray analysis. Crystal
data for 8: orthorhombic, space group Pbca; a ) 14.351 (2), b ) 11.932 (1), c ) 23.278 (2) Å; V ) 3986 (1)
Å3; Z ) 8; R ) 0.057 for 2615 reflections. Crystal data for 10: orthorhombic, space group P212121; a ) 7.104
(1), b ) 12.805 (1), c ) 22.280 (2) Å; V ) 2026.7 (6) Å3; Z ) 4; R ) 0.066 for 3423 reflections. The rigid
ABIH2 ligand, whose chelating moieties are constrained to be cofacial, allows the preparation of a new family of
cofacial bimetallic complexes (ABI)[ML2]2 with controllable environments around the metal centers. Two novel
cofacial binuclear complexes 4 and 5, with ML2 ) dicarbonylrhodium(I) and (η4-1,5-cyclooctadiene)iridium(I),
have been synthesized by reaction of ABIH2 with [(µ-Cl)Rh(CO)2]2 and [(µ-Cl)Ir(COD)]2, respectively. NMR
data indicate the formation of meso and racemic atropisomers for 2a, 4, and 5.
Introduction
Chart 1
We recently studied cofacial binuclear transition-metal com-
plexes such as 11,2 (see Chart 1) as potential catalysts for
multielectron redox reactions. The cavities in these complexes
(M‚‚‚M 4.5-5.0 Å) may be appropriate for binding of small
guest molecules. Previous work on the bis(â-keto enamine)
complexes 1b2 demonstrated multielectron redox activity that
is not observed in the analogous mononuclear systems, sug-
gesting that the rigid structure improves the stability of the
complexes when they are oxidized. However, the need for two
bridging ligands, to enforce the desired cofacial geometry,
limited the complexes to M(diketonato)2 (1a) and M(imino
ketonato)2 (1b) coordination environments. More conforma-
tionally rigid ligands based on 1,8-disubstituted anthracenes (2),
with their chelating moieties constrained to be cofacial, would
eliminate the need for a second bridging group. Collman3a and
Chang3b have demonstrated this principle in their studies of
cofacial anthracene- and biphenylene-bridged diporphyrins;
complexes of these ligands are electrocatalysts for reduction of
O2. Our goal was to use the ligands 2 as an entry to bimetallic
complexes 3 with readily controllable environments around the
metal centers. These complexes should provide for a variety
of molecular and electronic structures by allowing the use of
many different ancillary ligands along with the anthracenediyl-
bridged ligands 2. The more open structure of these complexes
may also offer better access of substrate molecules to the
intramolecular binding site. We now report the preparation of
2a, ABIH2, and two of its cofacial complexes 3a, with ML2 )
X Abstract published in AdVance ACS Abstracts, November 1, 1997.
(1) (a) Maverick, A. W.; Klavetter, F. E. Inorg. Chem. 1984, 23, 4129.
(b) Maverick, A. W.; Martone, D. P.; Bradbury, J. R.; Nelson, J. E.
Polyhedron 1989, 8, 1549.
(2) Bradbury, J. R.; Hampton, J. L.; Martone, D. P.; Maverick, A. W.
Inorg. Chem. 1989, 28, 2392.
(3) (a) Collman, J. P.; Wagenknecht, P. S.; Hutchinson, J. E. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1537. (b) Liu, H. Y.; Abdalmuhdi, I.;
Chang, C. K.; Anson, F. C. J. Phys. Chem. 1985, 84, 666 and
references therein.
dicarbonylrhodium(I) ((ABI)[Rh(CO)2]2, 4) and (η4-1,5-cy-
clooctadiene)iridium(I) ((ABI)[Ir(COD)]2, 5).
The SN2 synthetic strategy normally used for alkylation of
acetylacetonates cannot generally be used to prepare 3-aryl-
acetylacetones such as 2b. One method to activate aryl halides
toward nucleophilic substitution is π-complexation to Fe; such
S0020-1669(97)00635-6 CCC: $14.00 © 1997 American Chemical Society

A Binucleating Ligand Based on Anthracene
Inorganic Chemistry, Vol. 36, No. 25, 1997 5827
Scheme 1
Scheme 2
an approach has provided a route to 3-phenylacetylacetone.4
We attempted to prepare ligands 2 by this method; although
we succeeded in preparing a diiron complex from 1,8-dichlo-
roanthracene,5 we were unable to convert it to 2b. As an
alternative strategy for 2, we chose the 3,5-dimethylisoxazol-
4-yl (DMI) group as an acetylacetone synthon. Isoxazoles can
be reductively cleaved to form â-keto enamines, which can
usually be hydrolyzed to the corresponding â-diketones.6 Our
successful preparation of 2a involved palladium-catalyzed cross-
coupling between 4-(trialkylstannyl)-3,5-dimethylisoxazole and
1,8-dibromoanthraquinone or 1,8-dibromoanthracene. Direct
complexation of [(µ-Cl)Rh(CO)2]2 and [(µ-Cl)Ir(COD)]2 with
2a in the presence of a base yielded the cofacial binuclear
complexes 4 and 5, respectively.
Results and Discussion
The Binucleating Ligand. We first attempted to cross-
couple 1,8-disubstituted anthracene electrophiles (e.g., 1,8-
dichloroanthracene and 1,8-ditriflatoanthraquinone7) with Grig-
nard8 and organozinc9 derivatives of DMI, but their high
viscosity made them difficult to handle. Therefore, we searched
for a more stable and manageable DMI organometallic com-
pound. Treatment of the Grignard reagent of DMI with R3-
SnCl (R ) Me, Bu) in THF afforded the trimethyl- (6a) and
tributyltin (6b) reagents of DMI as air-stable, pale yellow liquids
(Scheme 1). This method provides a good alternative to the
literature procedure, in which 6a and 6b were obtained via
DMI-organolithium reagents in 69%10 and 51%6b yields,
respectively.
1,8-dichloroanthraquinone with excess KBr.12 Compound 7
couples with 6a under Pd catalysis (Stille reaction) to produce
1,8-bis(3,5-dimethylisoxazol-4-yl)anthraquinone (8) (Scheme 2).
Reduction of 8 with Zn in NH3(aq)13 afforded 1,8-bis(3,5-
dimethyl-4-isoxazolyl)anthracene (10) in good yield. Lower
overall yields of 10 were obtained by reaction of 1,8-dibro-
moanthracene (9)14 with the tin reagent 6b. In addition, the
anthraquinone 8 is easier to purify by column chromatography
than the anthracene derivative 10, owing to its higher polarity
in relation to the other reaction side products; 10 usually elutes
as a mixture with the side product 3,5,3′,5′-tetramethyl-4,4′-
biisoxazole.
Methyl coupling also occurred during the conversion of 7 to
8, forming 1-(3,5-dimethylisoxazol-4-yl)-8-methylanthraquinone
in 12% yield. The generally faster rate of methyl transfer as
well as the use of the more hindered anthraquinone 7 may
explain the formation of some alkyl transfer product; in the
anthracene system (9 + 6b), on the other hand, butyl coupling
went undetected.
On the basis of the method described for the 1,5-isomer,11
we prepared 1,8-dibromoanthraquinone (7) (as a ca. 10:1
mixture with 1-bromo-8-chloroanthraquinone) by reaction of
(4) Sutherland, R. G.; Abd-El-Aziz, A. S.; Pio´rko, A.; Lee, C. C. Synth.
Commun. 1987, 17, 393.
(5) Benites, M. R.; Fronczek, F. R.; Maverick, A. W. J. Organomet. Chem.
1996, 516, 17.
(6) (a) For a review, see: Gru¨nanger, P.; Vita Finzi, P. Isoxazoles; The
Chemistry of Heterocyclic Compounds, Vol. 49; Taylor, E. C.,
Weissberger, A., Eds.; John Wiley & Sons, Inc.: New York, 1991;
Part 1, pp 273-281 and 391-416. (b) Sakamoto, T.; Kondo, Y.;
Uchiyama, D.; Yamanaka, H. Tetrahedron 1991, 47, 5111. (c) Alberts,
A. H.; Cram, D. J. J. Am. Chem. Soc. 1979, 101, 3545.
(7) Benites, M. R.; Fronczek, F. R.; Maverick, A. W. Unpublished work.
(8) The Grignard reagent of 4-iodo-3,5-dimethylisoxazole (Kochetkov,
N. K.; Sokolov, S. D.; Vagurtova, N. M. J. Gen. Chem. USSR (Engl.
Transl.) 1961, 31, 2167) is obtained as a suspension when prepared
in diethyl ether/THF (Kochetkov, N. K.; Sokolov, S. D. J. Gen. Chem.
USSR (Engl. Transl.) 1963, 33, 1174) or pure THF or as a syrup in
diethyl ether.
(9) The Grignard reagent could be converted to an organozinc reagent by
treatment with anhydrous ZnCl2 in Et2O (see, for example: Methoden
der Organischen Chemie; Mu¨ller, E., Bayer, O., Meerwein, H., Ziegler,
K., Eds..; Thieme: Stuttgart, Germany, 1973; Vol. 13/ 2a, pp 650-
651), but this material was also too viscous for use in cross-coupling
reactions.
(10) Nesi, R.; Ricci, A.; Taddei, M.; Tedeschi, P. J. Organomet. Chem.
1980, 195, 275.
(11) (a)Hamanoue, K.; Kajiwara Y.; Miyake, T.; Nakayama, T.; Hirase,
S.; Teranishi, H. Chem. Phys. Lett. 1983, 94, 276. (b) Brienne, M.-J.;
Jacques, J. Bull. Soc. Chim. Fr. 1973, 190.
(12) This procedure has since been modified to give pure compound 7 in
39% overall yield (Holden, T. M.; Isovitsch, R. A.; Maverick, A. W.
Unpublished work, 1995): the mixture of 7 and 1-bromo-8-chloro-
anthraquinone produced in the first KBr treatment was crystallized
from hot nitrobenzene before the second treatment with KBr.
(13) House, H. O.; Hrabie, J. A.; VanDerveer, D. J. Org. Chem. 1986, 41,
921.
(14) 9 was obtained as a ca. 10:1 mixture of 1,8-dibromo- and 1-bromo-
8-chloroanthracenes, by starting from the corresponding 10:1 an-
thraquinone mixture and following the procedure reported by: Haenel,
M. W.; Jakubik, D.; Kruger, C.; Betz, P. Chem. Ber. 1991, 124, 333.

5828 Inorganic Chemistry, Vol. 36, No. 25, 1997
Table 1. 1H NMR Dataa
Benites et al.
compd
H-2
H-3
H-4
H-9 H-10 H-11 H-12
NH
L
8
8.41 (dd, 7.8, 1.2) 7.80 (t, 7.7)
7.48 (dd, 7.6, 1.3)
1.99 2.21
1.93* 2.15*
10
2ab
4b
8.10 (d, 8.6)
7.55 (dd, 8.5, 6.8) 7.32 (dd, 6.8, 0.9) 7.87
8.61 2.03 2.20
2.01* 2.18*
8.50 1.61 1.68 5.60, 10.72
1.55* 1.65*
8.46 1.69 1.74 7.71
1.65* 1.71* 7.55*
8.50 1.69 1.75 7.84
1.67* 1.74* 7.74*
7.34 (dd, 6.7, 1.2) 7.48 (dd, 8.5, 6.7) 8.00 (d, 8.5)
7.33* (dd, 6.7, 1.2) 7.46* (dd?)c
7.32 (d, 6.7)
8.40
7.48 (dd, 8.5, 6.7) 7.98 (d, 8.5)
7.81
7.79*
8.23
5d
7.26 (d, 6.6)
7.43 (dd, 8.6,6.8) 7.97 (d, 8.5)
4.31 (1H), 4.18 (3H), 3.60 (1H),
3.44 (3H), 2.36 (8H), 1.88 (8H)
a 300 MHz, unless otherwise noted; in CDCl3, δ/ppm (J/Hz in parentheses); assignments were made using NOE experiments; starred peaks
correspond to the minor atropisomer. b 200 MHz. c Exact nature of pattern could not be determined due to overlap with 7.48 ppm resonance. d 250
MHz; all COD peaks appear as broad multiplets.
Table 2. 13C NMR Dataa
compd C-1
C-2
C-3
C-4 C-4a C-9 C-9a C-10 C-11 C-12 C-13
C-14
C-15
L
8
10
133.5 138.9 127.7 133.1 134.4 164.4 131.2 183.3 10.6 11.5 158.8 115.9 164.3
130.9 128.8 125.3 128.6 132.0 121.6 128.1 128.0 10.4 11.4 159.5 114.5 166.1
159.4*
166.0*
2ab
4c
132.6 129.3 125.6 127.6 138.1 122.0 132.2 127.4 21.6 28.5 161.0 106.4 196.8
28.4* 160.0* 106.1* 196.6*
132.6 128.6 125.8 127.5 139.1 121.9 132.3 126.9 25.9 28.1 167.1 108.0 179.0 187.2 (d, 70.8), 187.1 (d, 71.2),
128.5* 126.8* 28.0* 166.8* 178.3* 186.1 (d, 64.2), 185.9 (d, 63.8)
132.3 130.2 125.5 127.8 139.4 122.0 128.0 132.3 30.6 32.0 164.6 108.6 178.1 66.8, 66.2, 66.0, 53.4, 52.9, 52.7,
30.5* 52.6, 28.3, 27.1
5d
a 63 MHz, unless otherwise noted; in CDCl3, δ/ppm; assignments for 2a and 4 were made using DEPT, HMQC, and HMBC experiments;
assignments for compounds 8, 10, and 5 were made by DEPT experiments and by comparison with the data from 2a and 4; starred peaks correspond
to the minor atropisomer. b 50 MHz. c 75 MHz; JC-Rh/Hz in parentheses. d 63 MHz.
Table 3. Crystallographic Data for Compounds 8 and 10
We expected that the new dimethylisoxazole derivatives 8
8
10
and 10 would show atropisomerism, with racemic and meso
atropisomers possible (meso shown in Scheme 2). Intercon-
version of the two atropisomers (e.g., by rotation of one of the
DMI moieties about the anthracene-DMI C-C bond) is likely
to be extremely slow in both compounds for steric reasons.
Indeed, the NMR spectra of both 8 and 10 show two sets of
signals for the DMI methyl atoms (see Tables 1 and 2).
Integration ratios, calculated from the two sets of methyl group
protons (H-11 and H-12), suggest that both compounds are
formed as roughly equal (ca. 0.9:1.0) mixtures of the two
atropisomers, although we could not determine which is which.
Both 8 and 10 crystallize, and we determined their crystal
structures. Crystallographic data are summarized in Table 3.
Since it is usually possible to distinguish N and O atoms by
X-ray methods, we were interested in whether the atropisomer
distribution in either compound could be determined from the
X-ray data. The molecules in both structures lie on general
positions, so that there is no crystallographic symmetry imposed
on either molecule. Still, we were unable to distinguish reliably
between N and O in either structure. In 8 (see Figure 1), the
atoms were modeled as half N and half O, and they are shown
as “NO1”, “NO2”, etc. Attempts to refine them as separate N
or O atoms led to displacement parameters that were always
somewhat larger for O than for N; also, the refined C-het-
eroatom distances are not sufficiently different to make a
distinction. In 10, on the other hand, our model involves
separate N and O atoms, and their arrangement in Figure 1
appears to favor the meso atropisomer. However, the refined
formula
fw
space group
a, Å
C24H18N2O4
398.4
Pbca
14.351 (2)
11.932 (1)
23.278 (2)
3986.2 (12)
8
C24H20N2O2
368.4
P212121
7.104 (1)
12.805 (1)
22.280 (2)
2026.7 (6)
4
b, Å
c, Å
V, Å3
Z
T, °C
λ, Å
24
1.541 84
1.328
7.1
0.057
25
1.541 84
1.207
5.8
0.066
F
calcd, g cm-3
µ, cm-1
Ra
Rw
b
0.065
0.061
a R ) ∑[|Fo| - |Fc|]/∑|Fo|. b Rw ) [∑w(|Fo| - |Fc|)2/∑w|Fo| ]1/2
.
2
displacement parameters for O are still slightly larger than those
for N, and this makes assignment of a single predominant
atropisomer difficult. Thus, the X-ray data for 8 and 10 do not
help in determining the atropisomer distribution, but they do
not conflict with the NMR data (which suggest that roughly
equal quantities of meso and racemic atropisomers are present
in both compounds).
Hydrogenolysis of 10 with Raney nickel gave the bis(â-keto
enamine) 2a (ABIH2) in good yield. Attempts to selectively
cleave the N-O bond in anthraquinone 8, using RaNi or Mo-
15
(CO)6 as the reducing agent, gave unidentified mixtures of
(15) Nitta, M.; Kobayashi, T. J. Chem. Soc., Chem. Commun. 1982, 877.

A Binucleating Ligand Based on Anthracene
Inorganic Chemistry, Vol. 36, No. 25, 1997 5829
presence of BaCO3. Kumar et al. have reported the formation
of dirhodium(I) and diiridium(I) complexes containing the
binucleating Schiff base R,R′-bis(salicylideneamino)-m-xylene
(SIXH2)19 by reaction of the corresponding [(µ-Cl)M(COD)]2
and SIXH2 with NaOH(aq). On the basis of this method, we
have successfully converted the new binucleating ligand 2a to
its cofacial dicarbonylrhodium (4) and (η4-1,5-cyclooctadiene)-
iridium (5) binuclear complexes, by reaction with [(µ-Cl)Rh-
(CO)2]2 and [(µ-Cl)Ir(COD)]2, respectively, in the presence of
K2CO3(aq) (Scheme 3). Treatment of 2a with [(µ-Cl)Rh(CO)2]2
and BaCO3(s), following the procedure of Rubailo et al.,17 also
afforded 4; however, no reaction took place when K2CO3(s)
was used as base under the same reaction conditions. Conver-
sion of the free ligand into its binuclear complexes was followed
1
by H NMR.
Both complexes were isolated as yellow, air-stable solids,
and they are soluble in organic solvents. Complex 5 decom-
poses in aerated solutions over a few hours, unlike 4, which
remains stable in solution for weeks.
1H and 13C NMR data, Tables 1 and 2, are consistent with
the proposed structures of 4 and 5. In general appearance, the
aromatic regions of the spectra of both bimetallic complexes
are similar to that of the free ligand 2a, with the exception of
the H-9 protons (located inside the cavity), which show upfield
shifts. A larger shielding effect is observed for H-9 in the
dirhodium complex. Complexation of 2a causes a general
downfield shift of the proton and carbon atoms of the â-keto
enamine moiety. On the other hand, the COD olefinic (3.4-
4.4 ppm) and aliphatic (1.8-2.4 ppm) protons are shifted upfield
in 5 compared to those of free COD (5.6 and 2.4 ppm,
respectively).
The spectral data support the formation of both meso and
racemic atropisomers for complex 4, as shown by the two sets
of resonances for H-9 and the amino and methyl (H-11 and
H-12) group protons, as well as for the carbons at the 2-, 10-,
12-, 13-, and 15-positions. Furthermore, the four doublets
shown by the carbonyl ligands in 4 (δ 187.2, 187.1 ppm (JC-Rh
≈ 71 Hz, trans to O)20 and 186.1, 185.9 ppm (JC-Rh ≈ 64 Hz,
trans to N)) are consistent with the presence of the two
geometries for the rhodium complex. NMR data for 5 do not
exhibit the same degree of splitting for the proton and carbon
resonances for the two isomeric forms as those for complex 4.
However, the presence of two sets of amino and methyl (H-11
and H-12) protons and methyl carbon C-11 resonances indicates
the formation of both atropisomers for the diiridium complex 5
in ca. 2.3:1 ratio (calculated from the integration of the imino
group protons), as observed for the free ligand 2a. The
proximity of other signals to the amino and methyl group proton
resonances of 4 prevents an accurate measurement of the ratio
of its atropisomers.
Figure 1. ORTEP diagrams of compounds 8 and 10.
products, possibly owing to partial reduction of the an-
thraquinone moiety under the reaction conditions. NMR spectra
for 2a (Tables 1 and 2) show several doubled H and 13C
1
resonances, which are consistent with the presence of two
atropisomers in ca. 2.3:1 ratio (calculated from the integration
of the two sets of methyl group protons, H-11 and H-12).
Surprisingly, 2a did not convert to the corresponding â-diketone
(2b, ABAH2) when treated with acid (aqueous HCl or H2SO4)6a
at room temperature or in refluxing EtOH. This unusual
resistance to hydrolysis may be caused by the bulky anthracene
substituent.
Metal Complexes. Although the chemistry of symmetrical
(â-keto enamine) complexes (such as 11) is well-known, there
have been few reports on the synthesis of the less symmetric
HRMS and FAB mass spectroscopies confirm the structures
of 4 and 5, respectively. The FAB mass spectrum of 5 (Figure
2) shows the parent ion at m/z 968-973; isotope peak intensities
are as expected. Infrared data for complex 4 show two strong
absorptions at 2073 and 2005 cm-1 for the CO ligands. The
carbonyl stretching for the â-keto enamine moiety shifts to lower
values on coordination of the 2a ligand (νCO 1608 cm-1), with
5 (νCO 1559 cm-1) showing a red shift that is 13 cm-1 greater
than that for 4 (νCO 1572 cm-1).
(â-imino ketonato)ML2 systems 12.16,17 Rh(I) and Ir(I) dicar-
bonyl complexes of Schiff base17 and other asymmetric N,O-
bidentate ligands18 have been synthesized by treatment of the
corresponding [(µ-Cl)M(CO)2]2 with the chelating ligand in the
(19) Kumar, R.; Fronczek, F. R.; Maverick, A. W. Abstracts of Papers,
200th National Meeting of the American Chemical Society, Wash-
ington, DC; American Chemical Society: Washington, DC, 1990;
INOR 297.
(20) Assignments were made by comparison to other dicarbonylrhodium
complexes of Schiff base ligands reported in the literature: Bresler,
L. S.; Buzina, N. A.; Varshavsky, Yu. S.; Kiseleva, N. V.; Cherkasova,
T. G. J. Organomet. Chem. 1979, 171, 229.
(16) Bonati, F.; Ugo, R. J. Organomet. Chem. 1967, 7, 167.
(17) Rubailo, A. I.; Selina, V. P.; Cherkasova, T. G.; Varshavskii, Yu. S.
Koord. Khim. 1991, 17, 530; Chem. Abstr. 1991, 115, 63383r.
(18) Ugo, R.; La Monica, G.; Cenini, S.; Bonati, F. J. Organomet. Chem.
1968, 11, 159.

5830 Inorganic Chemistry, Vol. 36, No. 25, 1997
Benites et al.
Bu3SnCl for 6b (3.85 mL, 13.6 mmol), the resulting yellow suspension
was stirred at room temperature overnight. The final reaction mixture
was quenched with H2O (100 mL) and the product extracted with diethyl
ether (3 × 20 mL). The organic phase was then dried over MgSO4,
filtered, and concentrated to give a brown crude oil. Purification by
flash chromatography and distillation under reduced pressure gave 6a
and 6b, respectively, as light yellow liquids. 1H NMR spectra of both
compounds agree with those reported in the literature.6b,10
6a: Rf 0.52 (hexane-EtOAc, 4:1); yield 2.48 g (59%); EI-MS m/z
(%) 261 (2, M+ for 120Sn), 246 (100, M+ - CH3), 216 (52, M+
-
3CH3), 165 (31, Sn(CH3)3), 135 (44, SnCH3), 120 (25), 80 (29, C5H6N),
43 (27).
Figure 2. Molecular ion portion of FAB-MS recorded for (ABI)[Ir-
(COD)]2 (5). The inset shows ion distribution calculated for 12,13C and
191,193Ir isotopes.
6b: bp 135-141 °C/0.6 mmHg (lit.6b 140-150 °C/0.6 mmHg); yield
4.0 g (63%); 13C NMR (CDCl3, 63 MHz) δ 164.7 (CN), 173.5 (CO),
105.7, 29.0, 27.2 (CH2), 13.6, 13.3, 13.2 (CH3), 9.6 (CH2); EI-MS m/z
(%): 386 (18, [M - H]+ for 120Sn), 330 (13, M+ - Bu), 274 (78, M+
- 2Bu + H), 218 (94, M+ - 3Bu + 2H), 216 (100, M+ - 3Bu), 121
(36), 80 (54, C5H6N), 43 (33), 41 (40).
Scheme 3
1,8-Dibromoanthraquinone (7). According to the procedure
reported for the 1,5- isomer,11 1,8-dichloroanthraquinone (2.0 g, 7.0
mmol) was treated with KBr (4.0 g, 33.6 mmol), CuCl2 (0.1 g, 0.7
mmol), and 85% H3PO4 (4 mL) in nitrobenzene (15 mL). Water was
distilled from the reaction mixture until the temperature reached 200
°C, and then the mixture was refluxed for 24 h. The crude product
was precipitated from the cooled mixture with methanol, collected, taken
up in CH2Cl2, and isolated by evaporation of the solvent. Purification
by column chromatography (CH2Cl2) yielded 2.0 g of a 6:1 mixture
1
(as estimated by H NMR) of 5 and 1-bromo-8-chloroanthraquinone.
The same procedure was repeated using this product as starting material
to give 1.3 g of a 10:1 mixture of 7 (yield ∼45%)12 and 1-bromo-8-
chloroanthraquinone. Spectral data for 7 agree with those reported in
the literature.22
7: 13C NMR (CDCl3, 50 MHz) δ 181.7, 181.4, 141.1, 137.7, 135.0,
133.4, 126.7, 122.0.
1-Bromo-8-chloroanthraquinone: EI-MS m/z (%) 324 (20), 322
(78), 320 (59, M+ for 79Br and 35Cl), 296 (5), 294 (22), 292 (18, M+
- CO), 268 (7), 266 (30), 264 (23, M+ - 2CO), 243 (6), 241 (17, M+
- Br), 215 (6), 213 (16, M+ - Br - CO), 185 (43), 150 (100), 75
(98), 74 (51).
Summary
We have prepared the new binucleating ligand 2a, via bis-
(3,5-dimethylisoxazol-4-yl) intermediates, and its cofacial bis-
(dicarbonylrhodium) (4) and bis((η4-1,5-cyclooctadiene)iridum)
(5) complexes. These compounds represent the first members
of a new family of cofacial bimetallic complexes. We expect
that these complexes will be available with a variety of metal
coordination environments and provide access for intramolecular
binding of small substrate molecules between the metal centers.
1,8-Bis(3,5-dimethylisoxazol-4-yl)anthraquinone (8). To a solu-
tion of 7 (0.68 g, 1.9 mmol) and 6a (1.7 g, 6.6 mmol) in toluene (45
mL) was added PdCl2(PPh3)2 (0.03 g, 0.04 mmol), and the reaction
mixture was brought to reflux under N2. After 72 h, the resulting brown
mixture was allowed to cool to room temperature, diluted with diethyl
ether (30 mL), and extracted with H2O (2% 15 mL). The organic phase
was then dried over MgSO4 and concentrated to give a crude orange
solid. Purification by flash chromatography afforded 8 and the major
side product 1-(3,5-dimethylisoxazol-4-yl)-8-methylanthraquinone as
yellow-orange solids. Crystals of 8 were obtained from hot i-PrOH.
8: Rf 0.23 (hexane-EtOAc, 1:1); yield 0.35 g (47%); EI-MS m/z
(%) 398 (47, M+), 357 (86, M+ - C2H3N), 338 (14, M+ - 4CH3), 324
(36), 296 (41), 286 (40), 272 (53), 214 (40), 200 (26), 189 (100), 174
(36), 75 (24).
Experimental Section
3,5-Dimethyl-4-iodoisoxazole9 and PdCl2(PPh3)221 were prepared by
following the literature procedures. Anhydrous THF and diethyl ether
were purchased from Aldrich Chemical Co. Other chemicals and
solvents were reagent grade and were used as received. Flash
chromatography was performed on silica gel 60 (230-400 mesh). NMR
spectra were recorded by using Bruker AC200, AC250, and ARX300
NMR spectrometers. Infrared spectra were obtained with a Perkin-
Elmer 1760X instrument. An HP 5971 instrument was used for GC-
MS. FAB and high-resolution mass spectra were acquired with
Finnigan TSQ-70 triple-quadrupole and Finnigan MAT-900 double-
focusing mass spectrometers, respectively, using 3-nitrobenzyl alcohol
as the matrix.
Trialkyl(3,5-dimethylisoxazol-4-yl)tins 6a and 6b. According to
the method described by Kochetkov and Sokolov,8 a mixture of 3,5-
dimethyl-4-iodoisoxazole (3.64 g, 16.3 mmol) and Mg turnings (1.3 g,
52 mmol) in refluxing THF (60 mL) was treated with a solution of
BrCH2CH2Br (2.81 mL, 32.6 mmol) in THF (10 mL) under N2, and
the mixture was refluxed for an additional 2.5 h. The orange-brown
suspension was allowed to cool to room temperature, and after dropwise
addition of Me3SnCl for 6a (1 M in THF, 16.4 mL, 16.4 mmol) or
1-(3,5-Dimethylisoxazol-4-yl)-8-methylanthraquinone: Rf 0.51
(hexane-EtOAc, 2:1); yield 0.07 g (12%); 1H NMR (CDCl3, 250 MHz)
δ 8.37 (dd, J ) 7.8, 1.6, 1H), 8.20 (dd, J ) 7.8, 1.3, 1H), 7.77 (t, J )
7.6, 1H), 7.66-7.55 (m, 2H), 7.49 (dd, J ) 7.6, 1.6, 1H), 2.68 (s, 3H),
2.28 (s, 3H), 2.06 (s, 3H); 13C NMR (CDCl3, 63 MHz) δ 185.8, 183.7,
164.2 (CO), 159.0 (CN), 141.3, 138.4, 138.2, 134.5, 134.3, 134.1, 132.8
(higher intensity of this resonance suggests coincidental overlap of two
CH signals), 132.7, 130.9, 127.5, 125.6, 116.6, 22.7, 11.4, 10.6; EI-
MS m/z (%) 317 (9, M+), 276 (100, M+ - C2H3N), 261 (62, M+
-
C2H3N - CH3), 205 (23), 176 (46), 151 (18), 88 (9), 75 (7), 63 (8).
1,8-Bis(3,5-dimethylisoxazol-4-yl)anthracene (10). (a) Via Cross-
Coupling. To a solution of 9 (0.36 g, 1.1 mmol)14 and 6b (1.8 g, 4.7
mmol) in toluene (25 mL) was added PdCl2(PPh3)2 (0.03 g, 0.04 mmol),
and the resulting orange solution was heated to reflux for 62 h under
N2. The final dark brown solution was allowed to cool to room
temperature, diluted with diethyl ether (30 mL), and extracted with
(21) Heck, R. F. Palladium Reagents in Organic Synthesis; Academic
(22) House, H. O.; Koepsell, D. G.; Jaeger, W. J. Org. Chem. 1973, 38,
Press: London, 1985; p 18.
1167.

A Binucleating Ligand Based on Anthracene
Inorganic Chemistry, Vol. 36, No. 25, 1997 5831
H2O (3% 15 mL) and saturated aqueous KF solution (4% 15 mL). The
organic phase was filtered, dried over MgSO4, filtered, and concentrated
to give a yellow solid. The crude product was purified by flash
chromatography (hexane-AcOEt (3:1), Rf ) 0.39), to give 10 as a
pale yellow solid, and crystallized by slow evaporation from the eluting
solution as pale yellow needles: yield 0.12 g (31%); EI-MS m/z (%)
368 (100, M+), 311 (12, M+ - C2H3NO), 284 (25, M+ - C4H6NO),
283 (25), 268 (23), 242 (38), 241 (50), 240 (35), 213 (43), 202 (20).
Small amounts of the monocoupled compounds 1-bromo- and 1-chloro-
8-(3,5-dimethylisoxazol-4-yl)anthracenes (M+ m/z 351-353 and 307-
309, respectively), were also formed, most likely because the reaction
was incomplete and because some 1-bromo-8-chloroanthracene was
present in the starting material. However, because of their similar Rf
values, no attempt was made to separate them.
(b) Via Reduction of 8. A mixture of 8 (0.27 g, 0.67 mmol) and
Zn dust (2.0 g) in 28% aqueous NH3 (30 mL) was heated at 50-55 °C
for 7 h. The resulting red mixture was allowed to cool and was filtered;
the residue and filtrate were each extracted with CH2Cl2 and EtOAc.
The combined organic extracts were evaporated to dryness. The pale
yellow residue was then dissolved in a hot mixture of i-PrOH (35-40
mL) and 12 M HCl (1-2 mL) and the solution refluxed for 4 h. The
final yellow solution was concentrated to give 10 as a yellow solid:
yield 0.20 g (82%).
(10 mL) was added BaCO3(s) (8-10 mg, 0.04-0.05 mmol) under N2,
and the mixture was stirred vigorously at room temperature. After 15
min, 1H NMR of the reaction mixture showed complete conversion of
the free ligand 2a to complex 4. The resulting mixture was worked
up as described above to give 4 as a yellow solid.
3,3′-(1,8-Anthracenediyl)bis(4-imino-2-pentanonato-N,O)bis((η4-
1,5-cyclooctadiene)iridium) (5, (ABI)[Ir(COD)]2). To a solution
containing [(µ-Cl)Ir(COD)]2 (30 mg, 0.042 mmol) and 2a (15.0 mg,
0.042 mmol) in dry diethyl ether (15 mL) was added dropwise aqueous
K2CO3 solution (1.5 mL, 0.33 M) under N2; the mixture was stirred
vigorously at room temperature. After 0.5 h, 1H NMR showed complete
conversion of the free ligand 2a to complex 5. The reaction mixture
was then evaporated to dryness under a stream of N2. The greenish
yellow crude solid was washed with hexane and then dissolved in CHCl3
or CH2Cl2; the solution was filtered and concentrated to afford 5 as a
yellow solid: IR (KBr, cm-1) 3421 (w, NH), 2934 (w, CH), 2873 (w,
CH), 2828 (w, CH), 1614 (sh), 1559 (s, CO), 1472 (m), 1386 (m);
FAB-MS m/z 968.0-972.1 (M+; see Figure 2).
X-ray Analyses. Diffraction data for 8 and 10 were collected on
an Enraf-Nonius CAD4 diffractometer fitted with a Cu KR source and
a graphite monochromator, using the θ-2θ scan method. Final unit
cell constants were determined from the orientations of 25 centered
high-angle reflections. The intensities were corrected for absorption
using ψ-scan data for five reflections. The MolEN23 set of programs
was used for structure solution and refinement. Full details of the
crystal structure analyses of 8 and 10 are available in CIF format via
the Internet.
3,3′-(1,8-Anthracenediyl)bis(4-amino-3-penten-2-one) (2a, ABIH2).
A mixture of 10 (0.20 g, 0.54 mmol) and RaNi (0.1-0.2 g) in EtOH
(150 mL) was heated at reflux for 8 h. The final reaction mixture was
allowed to cool and was filtered through Celite, and the residue washed
with EtOAc. The combined filtrates were evaporated to dryness to
Acknowledgment is made to the donors of the Petroleum
Research Fund, administered by the American Chemical Society,
for partial support of this research. We are grateful to Yen-
Hsiang Liu for assistance with the X-ray crystal structure
determinations, to Dr. Tracy McCarley and Rolly Singh for mass
spectral analyses, and to Dr. Jose´ Castan˜eda for NMR support
and helpful discussions. We also thank the LEQSF and NSF
for financial support for the purchase of the 300 MHz NMR
spectrometer and for improvements to the X-ray Crystallography
Facility (Grant LEQSF(1996-97)-ENH-TR-10).
give 2a as an orange-brown solid: yield 0.16 g (80%); IR (KBr, cm-1
)
3381 (vs br, NH), 2929 (w, CH), 1265 (m, CN), 1608 (vs, CO); HRMS
(m/z) calcd for C24H25N2O2 [M + H]+ 373.1916, found 373.1917.
3,3′-(1,8-Anthracenediyl)bis(4-imino-2-pentanonato-N,O)bis(di-
carbonylrhodium) (4, (ABI)[Rh(CO)2]2). (a) Using K2CO3(aq) as
Base. To a solution containing [(µ-Cl)Rh(CO)2]2 (12 mg, 0.03 mmol)
and 2a (10.0 mg, 0.03 mmol) in dry diethyl ether (10 mL) was added
dropwise aqueous K2CO3 solution (0.6 mL, 0.2 M) under N2; the
1
mixture was stirred vigorously at room temperature. After 0.5 h, H
NMR of the reaction mixture showed complete conversion of the free
ligand 2a to complex 4. The resulting reaction mixture was then
evaporated to dryness under a stream of N2. The dark crude solid was
taken up in CHCl3 and the solution filtered and concentrated to afford
pure 4 as a yellow solid: IR (KBr, cm-1) 3451 (br, NH), 2071 (vs,
CO), 1996 (vs, CO), 1571 (m, CO), 1415 (m); IR (CHCl3, cm-1) 2073
(vs), 2005 (vs), 1572 (m, CO); HRMS (m/z) calcd for C28H23N2Rh2O6
[M + H]+ 688.9666, found 688.9615.
Supporting Information Available: 1H and 13C NMR spectra for
complexes 4 and 5 (2 pages). X-ray crystallographic files, in CIF
format, for compounds 8 and 10 are available on the Internet only.
Ordering and access information is given on any current masthead page.
IC9706358
(b) Using BaCO3(s) as Base. To a solution containing [(µ-Cl)Rh-
(CO)2]2 (12 mg, 0.03 mmol) and 2a (10.0 mg, 0.03 mmol) in benzene
(23) Fair, C. K. MolEN: An InteractiVe Structure Solution Procedure;
Enraf-Nonius: Delft, The Netherlands, 1990.