Solid- and solution-phase synthesis of a library of mixed-metal complexes

Article abstract of DOI:10.1002/ejic.200400244

A solid-phase synthesis protocol has been developed for the stepwise assembly of mixed-metal dinuclear complexes built from chromium, molybdenum and tungsten and a directional bridging ligand. Complexes with all nine possible metal combinations have been prepared on polystyrene/divinylbenzene copolymer employing a metal-complex-compatible silyl ether linker. Release of the metal complexes from the polymeric support is accomplished by fluoridolysis under mild conditions. The metal sequence in these complexes is precisely determined by the reaction sequence and is retained under the applied reaction conditions as shown by IR, NMR and UV/Vis spectroscopic analysis of the cleaved soluble products. Additionally, the same complex library has been prepared by conventional solution-phase synthesis to allow evaluation of the advantages and disadvantages of the solid-phase synthesis method. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200400244

FULL PAPER
Solid- and Solution-Phase Synthesis of a Library of Mixed-Metal Complexes
Katja Heinze*[a] and Juan D. Bueno Toro[a]
Keywords: Bridging ligands / Heterometallic complexes / Isomers / Group 6 elements / Solid-phase synthesis
A solid-phase synthesis protocol has been developed for the
stepwise assembly of mixed-metal dinuclear complexes built
from chromium, molybdenum and tungsten and a directional
cisely determined by the reaction sequence and is retained
under the applied reaction conditions as shown by IR, NMR
and UV/Vis spectroscopic analysis of the cleaved soluble
bridging ligand. Complexes with all nine possible metal com- products. Additionally, the same complex library has been
binations have been prepared on polystyrene/divinylben- prepared by conventional solution-phase synthesis to allow
zene copolymer employing a metal-complex-compatible silyl evaluation of the advantages and disadvantages of the solid-
ether linker. Release of the metal complexes from the poly-
meric support is accomplished by fluoridolysis under mild
phase synthesis method.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
conditions. The metal sequence in these complexes is pre- Germany, 2004)
Introduction
metalϪligand bonds of organometallic building blocks are
much more labile than typical bonds of organic molecules,
Monodisperse oligomers are essential for understanding and thus the required orthogonal stability and reactivity of
the chemistry and the physics of polymeric organic and in- the polymerϪmolecule bond and the connections between
organic materials.[1] Several synthetic strategies have been the building blocks is much more difficult to achieve for
developed for their synthesis — self-organisation processes, organometallic complexes than for organic molecules.
random syntheses followed by separation processes or step-
A system satisfying the above criteria consists of a silyl
by-step syntheses — all of which have advantages and dis- ether linker between the molecule and the polymer support,
advantages. Full control over chain length, end groups and which is cleavable by fluoride ions,[18] and inert
building-block sequence can be achieved using a stepwise metalϪisocyanide bonds between the organometallic build-
assembly of building blocks.
ing blocks.[19]
Although the step-by-step synthesis is straightforward, it
The solution and solid-phase synthesis of homometallic
becomes increasingly difficult for higher oligomers due to di- and trinuclear complexes with carbonylmolybdenum
the purification steps needed. An elegant solution to this complexes as building blocks has been reported recently.[5]
problem is synthesis on a solid support (solid-phase syn- The extension of this synthetic strategy to assembling
thesis[2]). This synthetic method, introduced by R. B. Merri- mixed-metal dinuclear complexes is reported in this contri-
field for the synthesis of peptides,[3] allows the purification bution.
of all intermediates by simple filtration. The development
of new synthetic sequences which enable the (preferably
combinatorial) solid-phase synthesis of non-natural, func-
tionalised oligomeric compounds might aid the identifi-
cation of synthetic oligomers with the capability of per-
forming functions similar to those of peptides and proteins,
for example catalysis, molecular recognition, signal trans-
duction or conversion of chemical into mechanical energy.[2]
Organometallic compounds, however, have only recently
been synthesised by this method,[4Ϫ17] and the solid-phase
synthesis of oligometallic complexes, in particular, remains
a challenge.[5,17] This is probably due to the fact that the
[a]
Anorganisch-Chemisches Institut der Universität Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg, Germany
Fax.: (internat.) ϩ 49-6221-545707
E-mail: katja.heinze@urz.uni-heidelberg.de
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Scheme 1
3498
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400244
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507

Synthesis of a Library of Mixed-Metal Complexes
FULL PAPER
All reactions in solution had to be performed under exact
stoichiometric conditions to reduce the formation of side
products such as XϪ(NʝNЈ)M(CO)4 (X ϭ CN, OH) or
CϵNϪ(NʝNЈ)M(CO)3CϵNϪ(NʝNЈ) which are difficult
to remove after the synthesis of the dinuclear complexes
(see Exp. Sect.).
Results and Discussion
General Solution- and Solid-Phase Synthesis
The hydroxy-functionalised Schiff-base ligand 1a was
used as starting point for the stepwise assembly of mixed-
metal complexes. For the solid-phase synthesis approach 1a
was attached to polystyrene/divinylbenzene copolymer by a
silyl ether linker to give the immobilised ligand 1b
(Scheme 1).[4] The silyl ether allows the release of the final
products of the solid-phase synthesis from the support by
mild fluoridolysis.[4,5,18]
Properties of the Soluble Complexes 5a؊16a
The two Schiff-base ligands present in the mononuclear
complexes 5aϪ7a give rise to several overlapping signals in
the aromatic region of the 1H NMR spectra (see Exp.
Sect.). Pronounced chemical-shift differences are observed
for the two imine protons H7/H7Ј and the two protons α to
the pyridine nitrogen atom (H12/H12Ј) due to the presence
(H7: δ ϭ 8.58Ϫ8.89 ppm; H12: δ ϭ 9.30Ϫ9.41 ppm) and
absence (H7Ј: δ ϭ 8.73Ϫ8.74 ppm; H12Ј: δ ϭ 8.53Ϫ8.59
ppm) of a coordinated metal centre (see Scheme 2 and Exp.
Sect. for atom numbering).
Complexes 5aϪ7a display the typical νCO absorption pat-
tern of a facially coordinated {M1(CO)3} fragment with dis-
turbed local C3v symmetry[4,19Ϫ21] and an absorption band
due to the coordinated isocyanide group of the
CϵNϪ(NʝNЈ) ligand in their IR spectra (see Exp. Sect.).
The absorption bands of 5aϪ7a in the UV/Vis spectra be-
tween 613Ϫ655 nm can be assigned to a metal-to-ligand
charge transfer transition of the {(diimine)M1(CO)3} chro-
mophore (see Exp. Sect.).[4,19Ϫ21] The MLCT bands show
negative solvatochromism as expected (see Supporting In-
formation).[22] A linear solvation-energy relationship analy-
sis reveals that the most dominant term influencing the
charge transfer is the polarity of the solvent, followed by
the hydrogen-acceptor ability of the solvent (see Support-
ing Information).[23]
Addition of a suitable {M1(CO)3} source (M1 ϭ Cr, Mo,
W) to 1a or 1b furnished the reactive soluble or polymer-
bound complexes 2aϪ4a and 2bϪ4b,[4], respectively
(Scheme 2). These THF-containing complexes were sub-
sequently treated with the bridging ligand (4-isocyanophen-
yl)(pyridin-2-ylmethylene)amine CϵNϪ(NʝNЈ)[19] to give
the stable tricarbonyl isocyanide complexes 5aϪ7a and
5bϪ7b, respectively (Scheme 2). The potentially chelating
diimine moiety of CϵNϪ(NʝNЈ) was finally coordinated
to an {M2(CO)4} fragment (M2 ϭ Cr, Mo, W) to give the
dinuclear mixed-metal complexes 8aϪ16a and 8bϪ16b,
respectively (Scheme 2). Each member of this ‘‘3 ϫ 3 li-
brary’’ of mixed-metal complexes has been prepared by
both solution- and solid-phase synthesis according to
Scheme 2.
The IR spectra of the dinuclear complexes are super-
positions of the spectra of the corresponding tricarbonyl
and tetracarbonyl complexes [disturbed local C3v and C2v
symmetry of the M(CO)n moieties, respectively];[4,19Ϫ21] one
νCN absorption and five νCO absorptions (two absorptions
are not resolved) are observed in each case (Table 1). This
finding was corroborated by DFT calculations of com-
plexes 8aϪ16a (see Supporting Information): the νCO and
νCN absorptions of the M1(CO)3(CNR) moieties are inde-
pendent of the metal in the M2(CO)4 fragment and vice
versa (Figure 1; all standard deviations are below 2 cmϪ1).
1
The H NMR spectra of 8aϪ16a consist of overlapping
signals in the aromatic region. Particularly informative is
the chemical shift of the two imine protons (H7/H7Ј: δ ϭ
8.35Ϫ8.90 ppm) and the two protons α to the pyridine ni-
trogen atom (H12/H12Ј: δ ϭ 9.14Ϫ9.40 ppm) which are in-
fluenced by metal coordination (cf. complexes 5aϪ7a,
Table 1, Exp. Sect.).
In the UV/Vis spectra of 8aϪ16a two absorptions are
observed in the visible region. These two absorptions arise
from MLCT transitions of the {(diimine)M1(CO)3} chrom-
ophore at lower energy and of the {(diimine)M2(CO)4}
chromophore at higher energy (Table 1).[4,19Ϫ21]
A thorough examination of the spectroscopic data reveals
that all dinuclear complexes exhibit similar but distinguish-
Scheme 2
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3499

K. Heinze, J. D. Bueno Toro
FULL PAPER
1
Table 1. Selected IR, UV/Vis and H NMR spectroscopic data of
8aϪ16a in THF
[a]
M2/M1
Cr
Cr
Mo
W
ν˜1CO
1830
1849
1904
1912
1826
1850
1906 (sh)
1917
2007
1827
1849
1903 (sh)
1911
2006
ν˜2CO
ν˜3CO
ν˜4CO
ν˜5CO
2007
ν˜CN
2073
2077
2073
λmax (ε)
δH(imine)
ν˜1CO
574 (4240)
8.51 (2 H)
1827
1849
1910
1913
2013
585 (7480)
8.35, 8.47
1824
1848
1904
1917
2012
600 (6070)
8.49, 8.90
1827
1848
1909 (br)
Mo
ν˜2CO
ν˜3CO
ν˜4CO
ν˜5CO
2013
ν˜CN
λmax (ε)
2070
2078
2067
546 (6925)
628 (5205)
8.54, 8.57
1825
1847
1894
536 (5575)
600 (4545)
8.57, 8.61
1824
1849
1894
547 (5290)
608 (4860)
8.57, 8.90
1826
1848
1895
δH(imine)
ν˜1CO
W
ν˜2CO
ν˜3CO
ν˜4CO
1915
2006
1918
2006
1912
2006
ν˜5CO
ν˜5CO
ν˜CN
λmax (ε)
2072
2075
2071
551 (6000)
643 (5210)
8.53, 8.87
554 (7425)
605 (5515)
8.57, 8.90
561 (8580)
606 (6860)
8.87, 8.90
Figure 2. IR spectra of complexes 8aϪ16a in THF
δH(imine)
[a]
IR values given in cmϪ1; λmax in nm, ε in Ϫ1·cmϪ1; δ in ppm.
(iv): one δH(imine) Ն 8.87 ppm: M1 or M2 ϭ W (10a, 13a,
14a, 15a);
(v): two δH(imine) Ն 8.87 ppm: M1 ϭ M2 ϭ W (16a).
With these criteria the sequences of all complexes, except
the Cr/W isomers 10a and 14a, can be unambiguously de-
termined, for example (i) and (iv) describe complex 15a.
The two isomeric complexes 10a and 14a can be dis-
tinguished either by their IR spectroscopic pattern around
1900 cmϪ1, as the difference between νCO and νCO amounts
to 8 cmϪ1 for 10a and 21 cmϪ1 for 14a (Figure 2, Table 1),
or by their MLCT absorptions in the UV/Vis spectra (Fig-
ure 3, Table 1). The energetic difference between the two
MLCT absorptions is smaller for 10a (2240 cmϪ1) than for
14a (3080 cmϪ1) as expected from the MLCT absorptions
of the corresponding mononuclear complexes (see Exp.
Sect. and refs.[4,21]).
Thus, all complexes can be distinguished by their spectro-
scopic signatures. These findings also imply that the com-
plexes retain their metal sequence under the applied reac-
tion conditions — there is no rearrangement, polymeris-
3
4
˜
˜
Figure 1. DFT (B3LYP, LANL2DZ) optimised geometry of com-
plex 8a (M1 ϭ M2 ϭ Cr) and standard deviations of the calculated
average IR wavenumbers [cmϪ1] of the M(CO)n fragments of com-
plexes 8aϪ16a
able features. This allows the determination of the metal ation or carbonyl scrambling.
sequence of the complexes simply by a combination of IR
In the solid state the IR data of 8aϪ16a show that hydro-
(Figure 2), 1H NMR and UV/Vis spectroscopic data gen bonds from the hydroxy group of the ligand 1a are pre-
(Table 1, Supporting Information).[19]
sent (νOH ϭ 3340Ϫ3469 cmϪ1, see Exp. Sect.). These ener-
˜
There are five obvious criteria to distinguish complexes gies are compatible with the formation of OH···O hydrogen
8aϪ16a:
bonds between OH groups of different complexes or be-
(i):νCN Ն 2075 cmϪ1: M1 ϭ Mo (9a, 12a, 15a) (Figure 2); tween the OH group and carbonyl ligands of different com-
˜
(ii): ν˜5CO Ն 2012 cmϪ1: M2 ϭ Mo (11a, 12a, 13a) (Fig- plexes.[20,21]
ure 2);
(iii): (i) and (ii): M1 ϭ M2 ϭ Mo (12a) (Figure 2);
For ligand 1a and complexes 5aϪ7a broad, structured
absorption bands are observed in the region 3000Ϫ3455
3500
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507

Synthesis of a Library of Mixed-Metal Complexes
FULL PAPER
Figure 3. UV/Vis spectra of 10a and 14a in THF and deconvol-
ution of the MLCT bands (Lorenz profiles)[24]
Figure 5. Structure of 7a in the solid state
tures (Table 2), although the intermolecular connectivity is
noticeable.
cmϪ1, which again suggests hydrogen-bonded structures.
This has been confirmed by determination of the crystal
structures of 1a and 7a (Figures 4 and 5). Intramolecular
bond lengths and angles are unremarkable for both struc-
In both solid-state structures the OH group of the ligand
forms hydrogen bonds to the nitrogen atom of a pyridine
ring of a neighbouring molecule (1a: O1···N2; 7a: O1···N5;
Table 3, Figures 4 and 5). The molecules of 1a form a zigzag
chain along the 001 direction (a glide plane of the space
˚
group Pbca; repeat distance c/2 ϭ 6.8 A), while the mol-
ecules of 7a form a linear chain along the 110 direction
˚
(repeat distance 13.9 A). The repeat distance for 7a is more
than twice as large as for 1a. Thus, 7a can be viewed as a
higher homologue of 1a with a W(CO)3ϪCϵNϪ(NʝNЈ)
fragment inserted between the hydrogen donor and ac-
ceptor groups. In addition, this insertion introduces a direc-
tional change and thus straightens the zigzag chain of 1a
(angle between OH vectors of adjacent molecules Ϯ61°) to
a linear chain of 7a (angle between OH vectors of adjacent
molecules 0°).
Properties of 5b؊16b
All polymer-bound complexes 5bϪ16b were directly
characterised on the solid support by IR and diffuse-reflec-
tance UV/Vis spectroscopy, as well as thermogravimetric
analyses (Table 4, Exp. Sect. and Supporting Information).
The progress of the heterogeneous reactions can be moni-
tored by IR spectroscopy due to the intense characteristic
carbonyl and isocyanide absorption bands in this otherwise
empty spectral region 1800Ϫ2100 cmϪ1 [5,25,26]
characteristic patterns allow the detection of {M(CO)3} and
.
These
Figure 4. Structure of 1a in the solid state
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3501

K. Heinze, J. D. Bueno Toro
FULL PAPER
˚
Table 2. Selected bond lengths [A] and angles [°] for 1a and 7a
Table 4. Analytical data of 8bϪ16b
[a]
1a
7a
M2/M1
Cr
Cr
Mo
W
[b]
[b]
W1ϪN1
W1ϪN2
W1ϪC13
W1ϪC14
W1ϪC15
W1ϪC16
C13ϪO13
C14ϪO14
C15ϪO15
C16ϪN3
N1ϪC7
N2ϪC8
C7ϪC8
C4ϪO1
O1ϪH1
N4ϪC23
2.212(8)
2.236(8)
1.968(11)
2.000(12)
1.940(10)
2.095(11)
1.159(11)
1.153(12)
1.179(12)
1.167(12)
1.281(12)
1.360(12)
1.475(14)
1.376(12)
1.12(11)
1.281(13)
1.322(13)
1.453(14)
72.2(3)
166.9(3)
94.0(4)
103.7(4)
87.0(3)
94.7(4)
96.8(4)
175.1(4)
89.8(3)
88.2(4)
89.3(4)
92.3(4)
ν˜1CO
ν˜2CO
1846
1890
1917
2008
2072
1826 (sh) 1850
1853
1922
2007
2081
3,4 [b]
CO
ν˜
1914
2008
2070
ν˜5CO
ν˜CN
λmax
[b]
[b]
[c]
420, 850 420, 730 420, 800
∆m/mexp (∆m/mcalcd. [d]) 10.9 (11.7) 11.5 (11.5) 9.0 (10.9)
[b]
Mo
W
ν˜1CO
ν˜2CO
1848
1888 (sh) 1851
1914
2015
2071
1822
1848
1890 (sh)
1911
2013
2077
[b]
3,4 [b]
CO
ν˜
1921
2013
2086
ν˜5CO
ν˜CN
λmax
[b]
[b]
[c]
420, 850 420, 800 420, 790
C4ϪO1
O1ϪH1
N1ϪC7
N2ϪC8
C7ϪC8
1.3593(13)
0.980(19)
1.2743(14)
1.3505(15)
1.4709(16)
∆m/mexp (∆m/mcalcd. [d]) 11.7 (11.5) 11.0 (11.2) 10.9 (10.6)
ν˜1CO
ν˜2CO
1847
1904
1918 (sh) 1918
2007
2073
1824
1903
1853
[b]
[b]
3,4 [b]
CO
N5ϪC24
C23ϪC24
ν˜
1918
2007
2075
ν˜5CO
ν˜CN
λmax
2007
2083
[b]
[b]
[c]
N1ϪW1ϪN2
N1ϪW1ϪC13
N1ϪW1ϪC14
N1ϪW1ϪC15
N1ϪW1ϪC16
N2ϪW1ϪC13
N2ϪW1ϪC14
N2ϪW1ϪC15
N2ϪW1ϪC16
C13ϪW1ϪC14
C13ϪW1ϪC15
C13ϪW1ϪC16
C14ϪW1ϪC15
C14ϪW1ϪC16
C15ϪW1ϪC16
N1ϪC7ϪC8ϪN2
N4ϪC23ϪC24ϪN5
C7ϪN1ϪC1ϪC2
C23ϪN4ϪC20ϪC21
420, 890 420, 670 420, 940
∆m/mexp (∆m/mcalcd. [d]) 11.2 (10.9) 11.5 (10.6) 10.4 (10.2)
IR values given in cmϪ1; λmax in nm, ε in Ϫ1·cmϪ1; δ in ppm;
[a]
[b]
[c]
∆m/m in %.
200 °C; ∆m/mcalcd. ϭ 100 ϫ (7 ϫ MCO)/[Mresin ϩ Mligand ϩ (7 ϫ
MCO) ϩ MM1 ϩ MM2] with MCO ϭ 28 g·molϪ1, Mresin
In CsI. In polytetrafluoroethylene. [d]∆m/mexp at
ϭ
,
1160 g·molϪ1 (resin loading 0.9 mmol·gϪ1), Mligand ϭ 207 g·molϪ1
MM1, MM2 ϭ 52 (Cr), 96 (Mo), 184 (W) g·molϪ1
.
86.2(4)
173.4(4)
87.3(4)
Ϫ2.6(13)
174.8(11)
131.3(11)
32.8(17)
(i): νCN Ն 2081 cmϪ1: M1 ϭ Mo (9b, 12b, 15b) (Table 4);
(ii): ν˜5CO Ն 2012 cmϪ1: M2 ϭ Mo (11b, 12b, 13b)
˜
N1ϪC7ϪC8ϪN2
C7ϪN1ϪC1ϪC2
Ϫ163.06(10)
Ϫ148.66(11)
(Table 4);
(iii): (i) and (ii): M1 ϭ M2 ϭ Mo (12b) (Table 4).
The diffuse-reflectance spectra of the polymer-bound di-
nuclear complexes 8bϪ16b show broad absorptions in the
visible spectral region which can be assigned to the MLCT
transitions of the two {(diimine)M(CO)n} chromophores
(n ϭ 3 at lower energy; n ϭ 4 at higher energy; Table 4).[4,6]
These absorption bands are considerably shifted to lower
energy relative to the MLCT bands of the soluble com-
plexes as previously observed for similar immobilised
complexes[4Ϫ6] (Tables 1 and 4). As the energy of the MLCT
absorptions is dominated by the solvent polarity (see above
and ref.[22]) the shift of the absorption bands of the immobi-
lised complexes towards lower energy should be due to
lower polarity of the polymer matrix. Even the degree of
˚
Table 3. Hydrogen-bonding geometry (distances in A, angles in °)
DϪH···A
DϪH
D···A
H···A
DϪH···A
1a: O1ϪH1···N2[a] 0.980(19) 2.7450(13) 1.772(20) 171.8(17)
7a: O1ϪH1···N5[b] 1.12(11)
2.757(12)
1.74(11)
148(9)
[a]
[b]
Symmetry code Ϫx ϩ 1/2, y, z ϩ 1/2. Symmetry code x ϩ 1,
y ϩ 1, z.
{M(CO)4} fragments on the polymer support. The intensity cross-linking influences the energy of the MLCT absorption
of these absorption bands can be compared to the intensity bands. Resins with lower cross-linking (1% vs. 2% di-
of the νCH absorption bands of the polymer backbone as vinylbenzene) induce an even more pronounced bathoch-
internal standard, thus allowing a semi-quantitative reac- romic shift of the MLCT bands (Figure 6, Table 4 and
tion control.[5] The final immobilised dinuclear complexes ref.[5]). Obviously, the resin polarity decreases with decreas-
display signals due to the {M(CO)3} and {M(CO)4} frag- ing cross-linking. This lower polarity of the 1% cross-linked
ments and due to the isocyanide functional group (Table 4). resin might be induced by the lower internal pressure due
The IR criteria (see above) for the different metalϪmetal to the pressure dependence of the polarity.[22] Thus, the
sequences apply for the immobilised complexes as well, al- MLCT absorption bands of the carbonyl complexes can be
lowing the confirmation of the correct chain-growth even used to measure the resin polarity of different polymeric
on a solid support:
supports.
3502
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507

Synthesis of a Library of Mixed-Metal Complexes
FULL PAPER
Figure 6. DRS UV/Vis spectra of 12b (polystyrene with 1% and
2% divinylbenzene)
Scheme 3
Thermogravimetric analyses of resins 8bϪ16b show that
carbon monoxide is released up to 200 °C, as detected from
the IR spectrum of the gas above heated samples.[5] The
weight losses ∆m/m up to 200 °C of around 11% correspond
well to the calculated weight losses (seven carbon monoxide
molecules per immobilised complex) based on the initial li-
gand loading of the polymer (Figure 7, Table 4 and Sup-
porting Information). This finding confirms the quantitat-
ive formation of the dinuclear complexes on the polymer,
as the corresponding tricarbonyl complexes 5bϪ7b show a
weight loss, ∆m/m, of around 6% (three carbon monoxide
molecules per immobilised complex, Figure 7).
Protonation of 8a؊Ϫ16a؊ with acetic acid yields the neu-
tral complexes 8aϪ16a, which were characterised as above.
In all cases no spectroscopic evidence was observed for the
formation of side products, thus confirming the identity
and stability of the polymer-bound complexes.
Advantages/Disadvantages of the Synthetic Methods
The obvious drawback of the solution synthesis is the
difficulty of obtaining pure samples. The solution synthesis
requires exact stoichiometric reaction conditions to avoid
formation of side products (mainly monomeric tetracar-
bonyl complexes) and tedious purification procedures of
the products.
In the solid-phase synthesis approach, however, excess re-
agents can be used, driving the reactions to completion and
thus avoiding the formation of polymer-bound side prod-
ucts. The reaction progress on the solid phase can be moni-
tored by IR spectroscopy.[5,25,26] On the other hand the
solid-phase synthesis requires adaptation of all reaction
conditions and optimisation of the solid support. For the
synthesis of the soluble complexes 8aϪ16a identical reac-
tion conditions can be applied irrespective of the metal cen-
tre to be incorporated (see Exp. Sect.). The synthesis of the
polymer-bound complexes 8bϪ16b, however, requires dif-
ferent reaction conditions depending on the metal fragment
{M(CO)3} (M ϭ Cr, Mo, W) to be attached. The highly
reactive chromium derivatives must be handled at low tem-
peratures (Ϫ30 °C) to avoid formation of tetracarbonyl
Figure 7. Correlation of experimental and calculated weight losses
of resins 5bϪ16b
In the final step of the solid-phase synthesis the dinuclear complexes, which seems to be promoted by the polymer
complexes were released from the support by cleavage of the support, while the less reactive carbonyltungsten complexes
SiϪO bond with fluoride ions (Scheme 3).[4,5,18] The soluble must be activated at higher temperatures (60 °C); the solu-
anionic complexes 8a؊Ϫ16a؊ can be detected in the IR and tion reactions all proceed at room temperature (see Exp.
UV/Vis spectra of the cleavage solution (Table 5). All νCO Sect.). In addition, for reactions involving {M(CO)n} (M ϭ
absorption bands of 8a؊Ϫ16a؊ are shifted to lower energy, Cr and W; n ϭ 3, 4) the properties of the polymer support
while the MLCT absorption bands are hypsochromically have to be optimised. The degree of cross-linking (2% divi-
shifted, due to the negatively charged ligand, relative to the nylbenzene in the initial experiments with molybdenum[4,5]
)
neutral parent complexes 8aϪ16a.
has to be reduced to 1% divinylbenzene giving polymers
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3503

K. Heinze, J. D. Bueno Toro
FULL PAPER
Table 5. IR and UV/Vis spectroscopic data of 8a؊Ϫ16a؊ in THF
[a]
M2/M1
Cr
Cr
Mo
W
ν˜1CO
ν˜2CO
ν˜3CO
ν˜4CO
ν˜5CO
ν˜CN
λmax
ν˜1CO
ν˜2CO
ν˜3CO
ν˜4CO
ν˜5CO
ν˜CN
λmax
ν˜1CO
ν˜2CO
ν˜3CO
ν˜4CO
ν˜5CO
ν˜CN
λmax
1805
1830
1892
1905
2000
2088
460, 535, 600 (sh)
1810
1832
1805
1833
1894
1908
2002
2087
465, 540, 590 (sh)
1808
1835
1897
1912
2009
2094
475, 544, 600 (sh)
1804
1831
1814
1841
1897
2002
2084
433, 530, 600 (sh)
1803
1832
1880 (sh)
1897
2006
2084
468, 565, 600 (sh)
1809
1831
1885
1903
2000
Mo
W
1897
2006
2071
470, 560, 600 (sh)
1810
1830
1885
1905
1884
1909
2000
2084
2000
2084
468, 550, 600 (sh)
2079
467, 550 (sh), 600 (sh)
450 (sh), 500 (sh), 600 (sh)
[a]
IR values given in cmϪ1; λmax in nm.
vents were dried by standard methods and distilled under argon
prior to use. 1a[4,5], 1b[4,5] and CϵNϪ(NʝNЈ)[19] were prepared by
literature methods. Polystyrene resins cross-linked with 1% di-
vinylbenzene and with a ligand loading of 0.9 mmol·gϪ1 were used
unless noted otherwise. The acetonitrile complexes were prepared
from (cycloheptatriene)M(CO)3 and (norbonadiene)M(CO)4,[27,28]
respectively, by reaction with CH3CN (M ϭ Cr, Mo, W). All other
reagents were used as received from commercial sources. NMR
with better swelling properties and larger effective pore-
sizes. The better swelling properties of the low cross-linked
resins impart a lower internal pressure (as shown by the
‘‘solvent’’ dependence of the MLCT absorption bands) and
thus render the heterogeneous reactions faster and more
selective.
spectra were recorded with
a Bruker Avance DPX 200 at
Conclusion
200.15 MHz (1H) at 303 K; chemical shifts are quoted in ppm with
respect to residual solvent peaks as internal standards (CD2Cl2:
δ ϭ 5.32 ppm). IR spectra were recorded with a BioRad Excalibur
FTS 3000 spectrometer using CaF2 cells or CsI disks. UV/Vis/NIR
spectra were recorded with a PerkinϪElmer Lambda 19, 0.2-cm
cells (Hellma, Suprasil). DRS-UV/Vis spectra were measured with
the same instrument using the PerkinϪElmer integrating sphere
and poly(tetrafluoroethylene) as reference. Mass spectra were re-
corded with a Finnigan MAT 8400 spectrometer with a 4-nitro-
benzyl alcohol matrix (FAB). Elemental analyses were performed
by the microanalytical laboratory of the Organic Chemistry De-
partment, University of Heidelberg. Thermogravimetric measure-
ments were carried out with a Mettler TC 50, heating rate
10 K·minϪ1 under argon from 30Ϫ800 °C.
Dinuclear mixed-metal complexes with different metal
centres at well-defined positions have been synthesised in
solution and on a polymer support. The metal sequence of
the soluble complexes can be determined by a combination
of standard spectroscopic techniques and is retained under
the applied reaction conditions. The solution synthesis is
straightforward; however, the purification of the products
is rather difficult, making this approach less suitable for
longer-chain complexes. The solid-phase synthesis, al-
though requiring more reaction steps (ligand immobilis-
ation and product release) and differently optimised reac-
tion conditions for each different reaction step, is much
easier to accomplish. Furthermore, the solid-phase syn-
thesis approach can easily be extended to longer-chain com-
plexes by repeating the chain-growth reaction steps before
the capping reaction without having to worry about solu-
bility and purification of intermediates. This demonstrates
the use of solid-phase synthesis methods in organometallic
combinatorial chemistry.
Solution Syntheses of 5a؊7a: Freshly prepared [M(CO)3(CH3CN)3]
(M ϭ Cr, Mo, W) was added to a solution of 1a (198 mg, 1 mmol)
in THF (40 mL). The blue solution was stirred until IR control
indicated complete formation of the THF complexes 2a, 3a or 4a.
CϵNϪ(NʝNЈ) (207 mg, 1 mmol) was added as a solid and the
Experimental Section
General: Unless noted otherwise, all manipulations were carried
out under argon by means of standard Schlenk techniques. All sol-
3504
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507

Synthesis of a Library of Mixed-Metal Complexes
FULL PAPER
solution turned green immediately. The green solution was stirred
until IR control indicated complete formation of the isocyanide
complexes 5a, 6a or 7a. The complexes were either isolated by ad-
dition of petroleum ether (boiling range 40Ϫ60 °C) and filtration
of the precipitate or the solutions were used directly for the prep-
arations of 8a؊16a.
10a: C32H19CrN5O8W (837.4): calcd. C 45.90, H 2.29, N 8.36;
found C 45.49, H 3.43, N 7.90. IR (CsI): ν˜ ϭ 3399 (br., OH), 2088
1
(m, CN), 2009 (m, CO), 1904 (vs, CO), 1831 (br., s, CO) cmϪ1. H
NMR (CD2Cl2): δ ϭ 6.9Ϫ7.1 (m, 2 H, H3,5), 7.3Ϫ7.7 (m, 8 H,
H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.9Ϫ8.1 (m, 4 H, H9, H9Ј, H10, H10Ј),
3
8.49 (s, 1 H, H7Ј), 8.90 (s, 1 H, H7), 9.29 (d, JH,H ϭ 5.4 Hz, 1 H,
3
H12,12Ј), 9.40 (d, JH,H ϭ 5.8 Hz, 1 H, H12,12Ј) ppm.
5a: MS (FAB): m/z (%) ϭ 541 (8) [Mϩ], 457 (48) [Mϩ Ϫ 3 CO].
UV/Vis (THF): λmax (ε) ϭ 655 nm (6330 Ϫ1·cmϪ1). IR (THF):
11a: C32H19CrMoN5O8 (749.5): calcd. C 51.28, H 2.56, N 9.34;
found C 51.43, H 3.53, N 9.62. MS (FAB): m/z (%) ϭ 749 (5) [Mϩ],
665 (10) [Mϩ Ϫ 3 CO]. IR (CsI): ν˜ ϭ 3449 (br., OH), 2091 (m,
ν˜ ϭ 2071 (m, CN), 1915 (vs, CO), 1861 (s, CO), 1826 (s, CO) cmϪ1
.
IR (CsI): ν˜ ϭ 3455 (br., OH), 2085 (m, CN), 1907 (vs, CO), 1838
CN), 2014 (m, CO), 1909 (vs, CO), 1833 (br., s, CO) cmϪ1 1H
.
1
(s, CO), 1811 (s, CO) cmϪ1. H NMR (CD2Cl2): δ ϭ 7.0Ϫ7.6 (m,
NMR (CD2Cl2): δ ϭ 7.0Ϫ7.1 (m, 2 H, H3,5), 7.2Ϫ7.7 (m, 8 H,
H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.8Ϫ8.1 (m, 4 H, H9, H9Ј, H10, H10Ј),
10 H, H2,6, H2Ј,6Ј, H3,5, H3Ј,5Ј, H11, H11Ј), 7.8Ϫ8.0 (m, 3 H, H9,
3
H10, H10Ј), 8.23 (d, JH,H ϭ 7.0 Hz, 1 H, H9Ј), 8.53 (br. s, 1 H,
3
8.54, 8.57 (s, 1 H, H7, H7Ј), 9.20 (d, JH,H ϭ 5.8 Hz, 1 H, H12,12Ј),
3
H12Ј), 8.58 (br. s, 1 H, H7), 8.74 (br. s, 1 H, H7Ј), 9.41 (d, JH,H
ϭ
3
9.40 (d, JH,H ϭ 5.0 Hz, 1 H, H12,12Ј) ppm.
5.2 Hz, 1 H, H12) ppm. Satisfactory elemental analyses could not
be obtained due to the inclusion of variable amounts of solvent
molecules.
12a: C32H19Mo2N5O8 (793.4): calcd. C 48.44, H 2.41, N 8.83;
found C 48.17, H 3.55, N 8.47. MS (FAB): m/z ϭ 792 (5) [Mϩ
Ϫ
H], 380 (100) [Mϩ Ϫ CNϪ(NʝNЈ) Ϫ Mo Ϫ 4 CO]. IR (CsI): ν˜ ϭ
6a: MS (FAB): m/z (%) ϭ 587 (31) [Mϩ], 559 (100) [Mϩ Ϫ CO],
380 (95) [Mϩ Ϫ CNϪ(NʝNЈ)]. UV/Vis (THF): λmax (ε) ϭ 613 nm
(3475 Ϫ1·cmϪ1). IR (THF): ν˜ ϭ 2076 (m, CN), 1918 (vs, CO),
1855 (s, CO), 1824 (s, CO) cmϪ1. IR (CsI): ν˜ ϭ 3388 (br., OH),
2089 (m, CN), 1911 (vs, CO), 1833 (s, CO), 1808 (s, CO) cmϪ1. 1H
3340 (br., OH), 2088 (m, CN), 2013 (m, CO), 1913 (vs, CO), 1908
1
(vs, CO), 1833 (br., s, CO) cmϪ1. H NMR (CD2Cl2): δ ϭ 6.99 (d,
3JH,H ϭ 8.8 Hz, 2 H, H3,5), 7.3Ϫ7.8 (m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј
,
H11, H11Ј), 7.8Ϫ8.1 (m, 4 H, H9, H9Ј, H10, H10Ј), 8.57 (s, 1 H, H7Ј),
3
8.61 (s, 1 H, H7), 9.20 (d, JH,H ϭ 4.8 Hz, 1 H, H12,12Ј), 9.30 (d,
3
NMR (CD2Cl2): δ ϭ 6.97 (d, JH,H ϭ 8.8 Hz, 2 H, H3,5), 7.3Ϫ7.7
3JH,H ϭ 4.8 Hz, 1 H, H12,12Ј) ppm.
(m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.8Ϫ8.0 (m, 3 H, H9, H10,
3
H10Ј), 8.2Ϫ8.3 (m, 1 H, H9Ј), 8.59 (d, JH,H ϭ 5.4 Hz, 1 H, H12Ј),
13a: C32H19MoN5O8W (881.3): calcd. C 43.61, H 2.17, N 7.95;
found C 42.45, H 2.93, N 7.92. MS (FAB): m/z ϭ 881 (30) [Mϩ],
853 (38) [Mϩ Ϫ CO], 741 (45) [Mϩ Ϫ 5 CO]. IR (CsI): ν˜ ϭ 3396
(br., OH), 2085 (m, CN), 2013 (m, CO), 1904 (vs, CO), 1830 (br.,
3
8.66 (s, 1 H, H7), 8.73 (br. s, 1 H, H7Ј), 9.30 (d, JH,H ϭ 5.2 Hz, 1
H, H12) ppm. Satisfactory elemental analyses could not be obtained
due to the inclusion of variable amounts of solvent molecules.
s, CO) cmϪ1 1H NMR (CD2Cl2): δ ϭ 6.9Ϫ7.1 (m, 2 H, H3,5),
.
7a: UV/Vis (THF): λmax (ε) ϭ 627 nm (3225 Ϫ1·cmϪ1). IR (THF):
7.3Ϫ7.7 (m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.9Ϫ8.1 (m, 4 H,
ν˜ ϭ 2071 (m, CN), 1913 (vs, CO), 1848 (s, CO), 1825 (s, CO) cmϪ1
.
H9, H9Ј, H10, H10Ј), 8.57 (s, 1 H, H7Ј), 8.90 (s, 1 H, H7), 9.20 (d,
3
3JH,H ϭ 5.4 Hz, 1 H, H12,12Ј), 9.40 (d, JH,H ϭ 5.6 Hz, 1 H,
IR (CsI): ν˜ ϭ 3412 (br., OH), 2087 (m, CN), 1906 (vs, CO), 1831
(s, CO), 1807 (s, CO) cmϪ1 1H NMR (CD2Cl2): δ ϭ 7.01 (d,
.
H12,12Ј) ppm.
3JH,H ϭ 8.8 Hz, 2 H, H3,5), 7.3Ϫ7.7 (m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј
,
H11, H11Ј), 7.8Ϫ8.0 (m, 3 H, H9, H10, H10Ј), 8.2Ϫ8.3 (m, 1 H, H9Ј),
14a: C32H19CrN5O8W (837.4): calcd. C 45.90, H 2.29, N 8.36;
found C 44.78, H 3.32, N 7.71. MS (FAB): m/z ϭ 725 (15) [Mϩ
Ϫ
8.58 (br. s, 1 H, H12Ј), 8.74 (br. s, 1 H, H7Ј), 8.89 (br. s, 1 H, H7),
4 CO], 669 (8) [Mϩ Ϫ 6 CO], 641 (10) [Mϩ Ϫ 7 CO]. IR (CsI): ν˜ ϭ
3450 (br., OH), 2084 (m, CN), 2007 (m, CO), 1906 (vs, CO), 1831
(br., s, CO) cmϪ1. 1H NMR (CD2Cl2): δ ϭ 6.9Ϫ7.1 (m, 2 H, H3,5),
7.2Ϫ7.7 (m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.8Ϫ8.1 (m, 4 H,
H9, H9Ј, H10, H10Ј), 8.53 (s, 1 H, H7), 8.87 (s, 1 H, H7Ј), 9.31 (br.
s, 1 H, H12,12Ј), 9.40 (br. s, 1 H, H12,12Ј) ppm.
3
9.40 (d, JH,H ϭ 5.2 Hz, 1 H, H12) ppm. Satisfactory elemental
analyses could not be obtained due to the inclusion of variable
amounts of solvent molecules.
Solution
Syntheses
of
8a؊16a:
Freshly
prepared
[(CH3CN)2M(CO)4] (M ϭ Cr, Mo, W) (1 mmol) was added to a
solution of 5a, 6a or 7a (1 mmol) in THF. The dark solution was
stirred until IR control indicated complete formation of the di-
nuclear complex. Addition of petroleum ether (boiling range
40Ϫ60 °C) precipitated the product complexes, which were purified
by repeated washings with diethyl ether (removal of mononuclear
complexes) and recrystallisation from THF/diethyl ether.
15a: C32H19MoN5O8W (881.3): calcd. C 43.61, H 2.17, N 7.95;
found C 44.38, H 3.32, N 8.02. MS (FAB): m/z ϭ 853 (5) [Mϩ
Ϫ
CO], 797 (4) [Mϩ Ϫ 3 CO], 685 (5) [Mϩ Ϫ 7 CO]. IR (CsI): ν˜ ϭ
3393 (br., OH), 2093 (m, CN), 2009 (m, CO), 1910 (vs, CO), 1829
(br., s, CO) cmϪ1. 1H NMR (CD2Cl2): δ ϭ 7.02 (d, 3JH,H ϭ 8.6 Hz,
2 H, H3,5), 7.2Ϫ7.7 (m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.7Ϫ8.0
(m, 4 H, H9, H9Ј, H10, H10Ј), 8.57 (s, 1 H, H7), 8.90 (s, 1 H, H7Ј),
8a: C32H19Cr2N5O8 (705.5): calcd. C 54.48, H 2.71, N 9.93; found
C 53.32, H 3.98, N 9.38. MS (FAB): m/z (%) ϭ 705 (5) [Mϩ]. IR
(CsI): ν˜ ϭ 3469 (br., OH), 2082 (m, CN), 2008 (m, CO), 1906 (vs,
3
3
9.20 (d, JH,H ϭ 5.2 Hz), 1 H, H12,12Ј, 9.40 (d, JH,H ϭ 5.2 Hz, 1
H, H12,12Ј) ppm.
1
CO), 1833 (br., s, CO) cmϪ1. H NMR (CD2Cl2): δ ϭ 7.1Ϫ7.7 (m,
16a: C32H19N5O8W2 (969.2): calcd. C 39.66, H 1.98, N 7.23; found
C 40.77, H 2.69, N 7.98. MS (FAB): m/z ϭ 969 (15) [Mϩ]. IR (CsI):
ν˜ ϭ 3464 (br., OH), 2085 (m, CN), 2007 (m, CO), 1900 (vs, CO),
10 H, H2,6, H2Ј,6Ј, H3,5, H3Ј,5Ј, H11, H11Ј), 7.7Ϫ8.0 (m, 4 H, H9, H9Ј,
3
H10, H10Ј), 8.51 (br. s, 2 H, H7, H7Ј), 9.29 (d, JH,H ϭ 3.6 Hz, 1 H,
3
H12/12Ј), 9.40 (d, JH,H ϭ 3.8 Hz, 1 H, H12/12Ј) ppm.
1831 (br., s, CO) cmϪ1. 1H NMR (CD2Cl2): δ ϭ 7.02 (m, 2 H, H3,5
,
3JH,H ϭ 7.2 Hz), 7.2Ϫ7.7 (m, 8 H, H2,6, H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј),
7.9Ϫ8.1 (m, 4 H, H9, H9Ј, H10, H10Ј), 8.87, 8.90 (s, 1 H, H7, H7Ј),
9a: C32H19CrMoN5O8 (749.5): calcd. C 51.28, H 2.56, N 9.34;
found C 51.46, H 3.49, N 9.36. IR (CsI): ν˜ ϭ 3449 (br., OH), 2091
3
3
9.32 (d, JH,H ϭ 5.2 Hz, 1 H, H12,12Ј), 9.40 (d, JH,H ϭ 5.8 Hz, 1
1
(m, CN), 2009 (m, CO), 1910 (vs, CO), 1831 (br., s, CO) cmϪ1. H
H, H12,12Ј).
NMR (CD2Cl2): δ ϭ 6.88 (br. d, 2 H, H3,5), 7.1Ϫ7.5 (m, 8 H, H2,6
,
H2Ј,6Ј, H3Ј,5Ј, H11, H11Ј), 7.7Ϫ7.9 (m, 4 H, H9, H9Ј, H10, H10Ј), 8.35,
Solid-Phase
Synthesis
of
5b؊7b:
Freshly
prepared
8.47 (s, 1 H, H7, H7Ј), 9.14 (m, 2 H, H12,12Ј) ppm.
[M(CO)3(CH3CN)3] (M ϭ Cr, Mo, W) (0.5 mmol), dissolved in
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3505

K. Heinze, J. D. Bueno Toro
FULL PAPER
CH3CN (15 mL), was added to resin 1b (ca. 0.25 mmol), swollen
Cleavage Procedure: A solution of tetra-n-butylammonium fluoride
in toluene (15 mL). The suspension was stirred (M ϭ Cr: 20 min, TBAF·3H2O (2 equiv.) in THF was added to a suspension of the
Ϫ30 °C; M ϭ Mo: 20 min, 20 °C; M ϭ W: 60 min, 60 °C) and the appropriate resin in THF (10 mL). The suspension was stirred at
resin turned from yellow to blue. The solvents were removed by ambient temperature for 6 h. IR spectroscopy indicated the forma-
filtration and the polymer was quickly washed with toluene/
CH3CN (1:1) until IR spectroscopy of the solution indicated com-
plete removal of the excess [M(CO)3(CH3CN)3]. Due to the sensi-
tivity of the immobilised complex, the resin swollen in toluene was
immediately treated with the ligand CϵNϪ(NʝNЈ) (104 mg,
0.5 mmol), dissolved in THF (5 mL). The polymer turned green
immediately, and after 30 min, the resin was washed with THF un-
til IR spectroscopy indicated complete removal of the excess
CϵNϪ(NʝNЈ). The resin was dried under reduced pressure giving
a dark green powder or used directly for further reactions. No sol-
uble (diimine)Mo(CO)x complexes were observed by IR spec-
troscopy at any stage of the reaction.
tion of the anionic complexes 8a؊Ϫ16a؊. The polymer was filtered
and washed with THF until the wash solutions were colourless.
The combined filtrates were acidified with acetic acid, concentrated
under reduced pressure and the complexes were precipitated by ad-
dition of diethyl ether. Recrystallisation as above yielded the com-
plexes 8aϪ16a. After cleavage, mass spectrometry of the remaining
silyl fluoride polymer indicated complete removal of the com-
plexes.[18]
Crystallographic Structure Determinations: The measurements were
carried out with an EnrafϪNonius Kappa CCD diffractometer
using graphite monochromated Mo-Kα radiation. The data were
processed using the standard Nonius software.[29] All calculations
were performed using the SHELXT PLUS software package. Struc-
tures were solved using direct or Patterson methods with the
SHELXS-97 program and refined with the SHELXL-97 pro-
gram.[30] Graphical handling of the structural data during refine-
ment was performed using XMPA[31] and WinRay.[32] Atomic coor-
dinates and anisotropic thermal parameters of the non-hydrogen
atoms were refined by full-matrix least-squares calculations. Data
relating to the structure determinations are collected in Table 6.
CCDC-233938 (1a) and -233939 (7a) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or
from the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ 44-1223-
336033; E-mail: deposit@ccdc.cam.ac.uk].
5b: IR (CsI): ν˜ ϭ 2074 (m, CN), 1919 (s, CO), 1863 (m, CO), 1833
(m, CO) cmϪ1. UV/Vis: λmax ϭ 420 (sh), 880 (br.) nm. TG (200
°C): ∆m/m ϭ 6.4% (calcd. 5.8%).
6b: IR (CsI): ν˜ ϭ 2084 (m, CN), 1922 (s, CO), 1857 (m, CO), 1826
(m, CO) cmϪ1. UV/Vis: λmax ϭ 420 (sh), 810 (br.) nm. TG (200
°C): ∆m/m ϭ 6.2% (calcd. 5.4%).
7b: IR (CsI): ν˜ ϭ 2072 (m, CN), 1913 (s, CO), 1850 (m, CO), 1822
(m, CO) cmϪ1. UV/Vis: λmax ϭ 420 (sh), 830 (br.) nm. TG (200
°C): ∆m/m ϭ 5.9% (calcd. 5.1%).
Solid-Phase Synthesis of 8b؊16b: Freshly prepared [M(CO)4-
(CH3CN)2] (0.5 mmol) dissolved in THF (15 mL) was added to
resin 5b, 6b or 7b (ca. 0.25 mmol), swollen in THF (5 mL). After
stirring at 20 °C for 2 h, the resin was washed with THF until IR
spectroscopy indicated complete removal of the excess [M(CO)4-
(CH3CN)2]. The resin was dried under reduced pressure giving a
black powder. No soluble (diimine)M(CO)x complexes were ob-
served by IR spectroscopy at any stage of the reaction.
Computational Methods: Density functional calculations were car-
ried out with the Gaussian03/DFT[33] program series. The B3LYP
formulation of density functional theory was used employing the
LanL2DZ basis set.[33] Harmonic vibrational frequencies and IR
Table 6. X-ray crystallographic data of 1a and 7a
1a
7a
Empirical formula
Formula mass
Crystal dimension [mm]
Crystal system
C12H10N2O
198.22
C28H19WN5O4·1.6THF
673.34·1.6THF
0.15 ϫ 0.10 ϫ 0.08
triclinic
0.50 ϫ 0.20 ϫ 0.20
orthorhombic
Pbca (61)
15.611(3)
9.285(2)
13.678(3)
90
90
90
1982.6(7)
8
0.087
¯
Space group (no.)
P1 (2)
˚
a [A]
9.2510(19)
9.4420(19)
21.324(4)
84.46(3)
87.01(3)
83.45(3)
1840.2(6)
2
˚
b [A]
˚
c [A]
α [°]
β [°]
γ [°]
3
˚
Cell volume [A ]
Molecular units per cell
µ [mmϪ1
]
3.184
1.420
200
Density (calcd.) [g cmϪ3
T [K]
]
1.328
200
Scan range (2Θ) [°]
5.2Ϫ60.1
10
22451
4.4Ϫ55.2
20
11662
Scan speed [s·frameϪ1
Measured reflections
Unique reflections
]
2902
8162
Obs. reflections [I Ն 2σ(I)]
Parameters refined
2119
176
0.24/Ϫ0.19
R1 ϭ 4.2%
Rw ϭ 10.9%
4691
424
3.00/Ϫ1.21
R1 ϭ 7.6%
Rw ϭ 18.1%
Ϫ3
˚
Max./min. of residual electron density [e·A
]
Agreement factors
(F2 refinement)
3506
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507

Synthesis of a Library of Mixed-Metal Complexes
FULL PAPER
[22]
[23]
ˇ
A. Vlcek, Jr., Coord. Chem. Rev. 2002, 230, 225Ϫ242.
M. J. Kamlet, J.-L. M. Abboud, M. H. Abraham, R. W. Taft,
J. Org. Chem. 1983, 48, 2877Ϫ2887.
intensities were calculated by numerical second derivatives using
analytically calculated first derivatives. Frequencies are not scaled.
Lorenz profiles y ϭ 2A/π[w/(4(x Ϫ xc)2 ϩ w2)]. For 10a: xc1
ϭ
[24]
Acknowledgments
This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie. The permanent, generous
support from Prof. Dr. G. Huttner is gratefully acknowledged.
342.1, w1 ϭ 108.7, A1 ϭ 434.8; xc2 ϭ 551.5, w2 ϭ 161.4, A2 ϭ
122.4; xc3 ϭ 629.2, w3 ϭ 122.9, A3 ϭ 53.5; R2 ϭ 0.99964; for
14a: xc1 ϭ 342.4, w1 ϭ 113.3, A1 ϭ 272.0; xc2 ϭ 551.5, w2 ϭ
136.6, A2 ϭ 79.4; xc3 ϭ 664.2, w3 ϭ 116.6, A3 ϭ 30.0; R2
0.99952.
ϭ
[25]
[26]
B. Yan, Acc. Chem. Res. 1998, 31, 621Ϫ630.
B. Yan, H.-U. Gremlich, S. Moss, G. M. Coppola, Q. Sun, L.
Liu, J. Comb. Chem. 1999, 1, 46Ϫ54.
[1]
Electronic Materials: The Oligomer Approach (Eds.: K. Müllen,
G. Wegner), Wiley-VCH Weinheim, 1998.
F. Zaragoza Dörwald, Organic Synthesis on Solid Phase, Wiley-
[2]
[27]
[28]
[29]
[30]
M. A. Bennett, L. Pratt, G. Wilkinson, J. Chem. Soc. 1961,
VCH Weinheim, 2000.
[3]
2037Ϫ2044.
R. B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149Ϫ2154.
[4]
E. W. Abel, M. A. Bennett, R. Burton, G. Wilkinson, J. Chem.
Soc. 1958, 4559Ϫ4563.
R. Hooft, Collect, Data Collection Software, Nonius, The Ne-
therlands, 1998; http://www.nonius.com
G. M. Sheldrick, SHELXS-97, Program for Crystal Structure
Solution, University of Göttingen, Germany, 1997; http://
www.shelx.uni-ac.gwdg.de/shelx/index.html; G. M. Sheldrick,
SHELXL-97, Program for Crystal Structure Refinement, Uni-
versity of Göttingen, Germany, 1997; http://www.shelx.uni-
ac.gwdg.de/shelx/index.html; International Tables for X-ray
Crystallography, Kynoch Press, Birmingham, U. K., 1974, vol.
4.
L. Zsolnai, G. Huttner, XPMA, University of Heidelberg, Ger-
many, 1998; http://www.rzuser.uni-heidelberg.de/~il1/laszlo/
xpm.html
R. Soltek, G. Huttner, WinRay, University of Heidelberg, Ger-
many, 1999; http://www.uni-heidelberg.de/institute/fak12/AC/
K. Heinze, Chem. Eur. J. 2001, 7, 2922Ϫ2932.
[5]
K. Heinze, J. B. Toro, Angew. Chem. 2003, 115, 4671Ϫ4674;
Angew. Chem. Int. Ed. 2003, 42, 4533Ϫ4536.
[6]
K. Heinze, V. Jacob, C. Feige, Eur. J. Inorg. Chem., in press.
[7]
M. S. Robillard, A. R. P. M. Valentijn, N. J. Meeuwenoord, G.
A. van der Marel, J. H. van Boom, J. Reedijk, Angew. Chem.
2000, 112, 3226Ϫ3229; Angew. Chem. Int. Ed. 2000, 39,
3096Ϫ3099.
[8]
K. S. Schmidt, M. Boudvillain, A. Schwartz, G. A. van der
Marel, J. H. van Boom, J. Reedijk, B. Lippert, Chem. Eur. J.
2002, 8, 5566Ϫ5570.
M. S. Robillard, J. S. Leith, G. A. van der Marel, J. H. van
[9]
[31]
[32]
[33]
Boom, J. Reedijk, Eur. J. Inorg. Chem. 2003, 1529Ϫ1532.
[10]
D. R. van Staveren, N. Metzler-Nolte, Chem. Commun. 2002,
1406Ϫ1407.
[11]
K. S. Schmidt, D. V. Filippov, N. J. Meeuwenoord, G. A. van
der Marel, J. H. van Boom, B. Lippert, J. Reedijk, Angew.
Chem. 2000, 112, 383Ϫ385; Angew. Chem. Int. Ed. 2000, 39,
375Ϫ377.
N. E. Landbeater, E. L. Sharp, Organometallics 2003, 22,
4167Ϫ4169.
huttner/frame/frame soft.html
Ϫ
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tom-
asi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratch-
ian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J.
W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Sal-
vador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G.
Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W.
Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision B.03,
Gaussian, Inc., Pittsburgh, PA, 2003.
[12]
[13]
S. I. Khan, A. E. Beilstein, M. Sykora, G. D. Smith, X. Hu,
M. W. Grinstaff, Inorg. Chem. 1999, 38, 3922Ϫ3925.
[14]
A. E. Beilstein, M. W. Grinstaff, Chem. Commun. 2000,
509Ϫ510.
[15]
I. Vargas-Baca, D. Mitra, H. J. Zulyniak, J. Banerjee, H. F.
Sleiman, Angew. Chem. 2001, 113, 4765Ϫ4768, Angew. Chem.
Int. Ed. 2001, 40, 4629Ϫ4632.
J. F. Valliant, R. W. Riddoch, D. W. Hughes, D. G. Roe, T. K.
[16]
Fauconnier, J. R. Thornback, Inorg. Chim. Acta 2001, 325,
155Ϫ163.
[17]
S. van Zutphen, M. S. Robillard, G. A. van der Marel, H. S.
Overkleeft, H. den Dulk, J. Brouwer, J. Reedijk, Chem. Com-
mun. 2003, 634Ϫ635.
[18]
K. Heinze, U. Winterhalter, T. Jannack, Chem. Eur. J. 2000,
6, 4203Ϫ4210.
[19]
K. Heinze, V. Jacob, Eur. J. Inorg. Chem. 2003, 3918Ϫ3923.
[20]
K. Heinze, J. Chem. Soc., Dalton Trans. 2002, 540Ϫ547.
Received March 26, 2004
Early View Article
Published Online June 18, 2004
[21]
K. Heinze, V. Jacob, J. Chem. Soc., Dalton Trans. 2002,
2379Ϫ2385.
Eur. J. Inorg. Chem. 2004, 3498Ϫ3507
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3507